An Intelligent Drone-Assisted Framework for Re-al-Time Oil Spill Detection Using SAR Image Analysis and Computer Vision
Authors-T.Pravalika, M Karishma
Keyword-Oil Spill Detection, Drone Surveillance, Synthetic Aperture Radar (SAR), Computer Vision, OpenCV, Image Processing, Real-Time Monitoring, Contour Detection, Marine Pollution, Environmental Monitoring.
Abstract-Oil spills pose a significant threat to marine biodiversity, coastal ecosystems, and maritime in-dustries, making rapid detection and monitoring essential for minimizing environmental damage. This research introduces an intelligent oil spill detection framework that integrates drone-based surveillance with Synthetic Aperture Radar (SAR) image analysis for continuous environmental monitoring. The proposed system processes live aerial video captured by drones together with uploaded SAR images using grayscale image conversion, threshold segmentation, contour ex-traction, and image processing techniques implemented through OpenCV. A graphical user interface allows users to switch between original and processed views for easier validation of detected oil spill regions. The framework estimates spill location and affected area while provid-ing real-time visualization to support rapid emergency response. Experimental evaluation demonstrates reliable detection performance under varying environmental conditions while main-taining low processing latency suitable for real-time applications. The combination of drone imaging and SAR-based analysis provides an efficient, scalable, and cost-effective solution for marine oil spill monitoring and environmental protection.
Doi-[https://doi.org/10.5281/zenodo.21678402]
A Comparative Machine Learning Framework for Speaker Identification Using Audio Biometric Fea-tures
Authors-B.Pradeep, Vijaya Mallamari
Keyword-Speaker Identification, Audio Biometrics, Machine Learning, Deep Learning, Support Vector Machine, Convolutional Neural Network, Long Short-Term Memory, Mel Spectrogram, Speech Recognition, VoxCeleb Dataset.
Abstract-Speaker identification has become an essential component of modern biometric authentication systems because voice characteristics provide a convenient and non-invasive method for verify-ing human identity. The rapid development automatic speaker identification. Audio recordings from the VoxCeleb dataset are preprocessed to eliminate unwanted noise and silence before extracting Mel Spectrogram features that effectively represent speech characteristics. The extract-ed features are used to train and evaluate each classification model using standard performance measures such as accuracy and F1-score. Experimental observations indicate that the SVM classifier delivers the highest recognition performance among the evaluated models, while CNN also achieves competitive results. In contrast, the LSTM model records comparatively lower performance because of the sequential complexity of the available dataset. The study demon-strates the importance of selecting an appropriate learning algorithm for speaker recognition applications and provides valuable insights for developing reliable and efficient voice-based biometric systems.
Doi-[https://doi.org/10.5281/zenodo.21678804]
Machine Learning-Based Active Wind Power Pre-diction Using Comparative Regression Models
Authors-Sai Bhanu Prasad, Pallavi Malloju
Keyword-Active Wind Power Prediction, Renewable Energy, Machine Learning, Linear Regression, Ridge Regression, Lasso Regression, K-Neighbors Regression, Decision Tree, Gradient Boost-ing, Principal Component Analysis, Wind Turbine, Energy Forecasting.
Abstract-The rapid expansion of renewable energy technologies has increased the need for accurate fore-casting techniques that can effectively estimate power generation from wind energy systems. Because wind conditions fluctuate continuously due to changing atmospheric and environmental factors, predicting the amount of electrical power generated by wind turbines remains a challeng-ing task. Reliable forecasting models are essential for improving grid stability, optimizing energy distribution, minimizing operational uncertainty, and supporting efficient utilization of renewable energy resources. In this research, a estimate active wind power by comparing the performance of multiple regression algorithms. The study employs six widely used supervised learning mod-els, namely Linear Regression, Ridge Regression, Lasso Regression, K-Neighbors Regression, Decision Tree Regression, and Gradient Boosting Regression. A real-world wind turbine dataset obtained from the Kaggle repository, consisting of meteorological measurements and turbine operational parameters recorded at regular intervals, is used for model development and evalua-tion. Before training the models, the dataset undergoes comprehensive preprocessing, including missing value estimation through imputation, detection and treatment of abnormal observations, feature correlation analysis, K-Neighbors Regression algorithm consistently delivers superior prediction accuracy compared with the remaining regression techniques, producing the lowest prediction errors and the highest coefficient of determination. The outcomes of this study con-firm that machine learning-based regression methods provide an effective solution for active wind power forecasting and can significantly contribute to intelligent energy management, im-proved scheduling of renewable power generation, and the reliable operation of modern smart grid systems.
Doi-[https://doi.org/10.5281/zenodo.21678954]
An Intelligent Machine Learning Framework for Real-Time Traffic Prediction and Traffic Flow Op-timization in Smart Urban Transportation Systems
Authors-S.Gouthami, M. Amulya
Keyword-Real-Time Traffic Prediction, Traffic Flow Optimization, Machine Learning, Random Forest, Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Intelligent Transportation Sys-tems, Smart City, Traffic Congestion Prediction, Cloud Computing, GPS-Based Traffic Moni-toring, Traffic Sensors, Route Optimization, Predictive Analytics, Urban Mobility.
Abstract-Rapid urban expansion and the continuous increase in the number of vehicles have created sig-nificant challenges for transportation authorities in maintaining smooth traffic movement and minimizing congestion. Conventional traffic control systems generally operate using fixed signal timings and predefined traffic rules, making them less effective in responding to dynamic traffic conditions caused by accidents, weather changes, road maintenance, public events, or unex-pected increases in vehicle density. These limitations often result in prolonged travel times, ex-cessive fuel consumption, increased environmental pollution, and reduced efficiency of urban transportation networks. To address these issues, this research presents an intelligent machine learning-based framework for real-time traffic prediction and traffic flow optimization that com-bines multiple real-time data sources with predictive analytics to support efficient traffic man-agement. The rewritten work preserves the methodology, machine learning algorithms, and experimental findings presented in the base paper while providing completely original academic content. The proposed framework acquires traffic information from various heterogeneous sources, including GPS-enabled vehicles, roadside traffic sensors, weather information services, public transportation systems, and cloud-based traffic databases. These data sources continuous-ly provide updated information regarding vehicle movement, traffic density, road conditions, weather variations, and travel patterns. Before model development, the collected dataset under-goes a comprehensive preprocessing phase involving data cleaning, duplicate removal, missing value handling, noise filtering, outlier detection, feature transformation, and normalization. Ex-ploratory Data Analysis (EDA) and correlation analysis are further performed to identify the relationships among traffic variables and determine the most influential features affecting con-gestion levels. These preprocessing operations improve dataset quality, reduce inconsistencies, and enhance the learning capability of the predictive models.To estimate future traffic conditions, the proposed system implements three supervised machine learning algorithms: Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbor (KNN). Random Forest constructs multiple decision trees to model complex traffic patterns and generate robust congestion predic-tions. Support Vector Machine performs accurate classification by identifying optimal decision boundaries between different traffic states, while K-Nearest Neighbor predicts traffic conditions by measuring the similarity between current observations and historical traffic patterns. Each algorithm is trained and evaluated independently using the same dataset to ensure a consistent and unbiased comparison of predictive performance. The trained models are then integrated with an intelligent traffic optimization module that dynamically recommends efficient travel routes and adaptive traffic signal strategies based on current congestion levels.
Doi-[https://doi.org/10.5281/zenodo.21679090]
Deep Learning-Based Automated Oral Cancer De-tection Using DenseNet169 and Transfer Learning for Early Clinical Diagnosis
Authors-M.Gouthami, Maragani Sony
Keyword-Oral Cancer Detection, Deep Learning, DenseNet169, LeNet, Transfer Learning, Convolutional Neural Network (CNN), Medical Image Classification, Image Augmentation, Computer-Aided Diagnosis, Oral Disease Recognition, Artificial Intelligence in Healthcare, ImageNet, Clinical Decision Support, Early Cancer Detection, Medical Image Analysis.
Abstract-Rapid urban expansion and the continuous increase in the nOral cancer is one of the most preva-lent and life-threatening diseases affecting the oral cavity, accounting for a significant number of cancer-related deaths worldwide. Despite considerable progress in medical diagnosis and treat-ment, the overall survival rate of oral cancer patients remains relatively low because the disease is frequently identified only during its advanced stages. Early diagnosis is therefore essential for improving treatment success, reducing disease progression, and increasing patient survival rates. Conventional diagnostic procedures mainly depend on visual examination, biopsy, histopatho-logical analysis, and the clinical expertise of healthcare professionals. Although these methods are considered reliable, they are often time-consuming, require experienced specialists, and may produce inconsistent results due to subjective interpretation. The rapid advancement of artificial intelligence and deep learning has created new opportunities for developing automated diagnostic systems capable of assisting clinicians in detecting oral cancer accurately and efficiently. The uploaded base paper presents an oral cancer detection framework using DenseNet169 and LeNet architectures with transfer learning, image augmentation, and comparative evaluation. This re-written research preserves the same methodology, algorithms, dataset structure, and experimental results while providing completely original academic content suitable for achieving a low plagia-rism score. The proposed research develops an intelligent computer-aided diagnostic framework for the automatic classification of oral diseases using deep learning techniques. The system is designed to analyze clinical images of the oral cavity and distinguish oral cancer from other common oral conditions, including healthy tongue, hairy tongue, leukoplakia, oral lichen, and oral thrush. A comprehensive image dataset containing multiple categories of oral conditions is utilized for model development. Before model training, all collected images undergo an extensive preprocessing procedure to improve image quality and maintain consistency across the dataset. Images are resized to a uniform resolution, normalized to standard pixel intensity ranges, and organized into appropriate class labels. To overcome the limitations of limited medical datasets and improve model generalization, several image augmentation techniques, including horizontal flipping, random rotation, zooming, and image shifting, are applied. These augmentation opera-tions increase dataset diversity, reduce overfitting, and improve the robustness of the deep learn-ing models during classification.The effectiveness of the proposed framework is assessed using several standard performance metrics widely employed in medical image classification research. These include Accuracy, Precision, Recall, and F1-Score, together with confusion matrix analy-sis to evaluate classification performance across all oral disease categories. Comparative experi-mental evaluation demonstrates that the DenseNet169 architecture substantially outperforms the LeNet model in every evaluation metric. The DenseNet169 model achieves an overall classifica-tion accuracy of 94.08%, precision of 94.16%, recall of 94.70%, and F1-score of 94.07%, indi-cating excellent diagnostic capability for multiclass oral disease recognition. In comparison, the LeNet architecture achieves comparatively lower classification performance, with an accuracy of 64.02%, precision of 64.06%, recall of 64.03%, and F1-score of 63.01%. The confusion matrix further confirms that DenseNet169 effectively distinguishes among multiple oral disease catego-ries while maintaining minimal classification errors, demonstrating its superior ability to learn complex visual representations from clinical images.
Doi-[https://doi.org/10.5281/zenodo.21679236]
An Intelligent Vision Transformer-Based Real-Time Driver Drowsiness Detection System for Ac-curate Eye State Classification and Road Safety Enhancement
Authors-S.Venkateswara Rao, Marupakula Shireesha
Keyword-Driver Drowsiness Detection, Vision Transformer (ViT), Deep Learning, Transfer Learning, Computer Vision, Eye State Classification, OpenCV, Haar Cascade, Image Processing, Real-Time Monitoring, Driver Fatigue, Road Safety, Artificial Intelligence.
Abstract-Driver drowsiness is a major factor contributing to road accidents and poses a serious threat to public safety. Continuous monitoring of a driver's alertness can significantly reduce fatigue-related accidents by providing timely warnings. This research proposes a real-time driver drows-iness detection system based on a Vision Transformer (ViT) model for accurate eye state classi-fication. The system utilizes a dataset containing approximately 84,900 images of open and closed eyes collected under different lighting and environmental conditions. Before training, the dataset undergoes preprocessing, label encoding, class balancing through random oversampling, and image augmentation techniques such as rotation, cropping, resizing, and sharpness adjust-ment to improve model robustness. The proposed framework employs the pre-trained Vision Transformer (ViT) model (google/vit-base-patch16-224-in21k) with transfer learning and addi-tional fully connected layers to enhance feature extraction and classification performance. The dataset is divided into 80% training, 10% validation, and 10% testing for effective model evalua-tion. The trained model is integrated with OpenCV and Haar Cascade face detection to perform real-time eye state recognition using a webcam. Whenever continuous eye closure is detected, an alarm is generated to alert the driver and prevent potential accidents. Experimental evaluation is performed using Accuracy, Precision, Recall, and F1-Score. The proposed system achieves an overall 98.8% accuracy, demonstrating its effectiveness in accurately identifying driver drowsi-ness while maintaining reliable real-time performance. The developed framework provides a practical, intelligent, and efficient solution for enhancing driver safety and reducing fatigue-related road accidents.
Doi-[https://doi.org/10.5281/zenodo.21721863]
A Secure Blockchain-Driven Framework for Smart Vehicle Procurement with Automated Con-tract Management and Transparent Ownership Transfer
Authors-P.Rupasri, Marka Sandhya
Keyword-Blockchain Technology, Smart Vehicle Procurement, Smart Contracts, Decentralized Ledger, Vehicle Ownership Transfer, Secure Transactions, Digital Identity Verification, Automotive Blockchain, Fraud Prevention, Distributed Ledger Technology.
Abstract-The increasing adoption of blockchain technology has transformed digital transaction systems by providing secure, decentralized, and transparent data management. The vehicle procurement process, however, still relies heavily on conventional procedures involving multiple intermediar-ies, manual documentation, and lengthy verification mechanisms that often increase operational costs and expose transactions to fraudulent activities. This paper presents a blockchain-enabled smart vehicle procurement framework that modernizes the complete purchasing lifecycle while preserving transaction integrity and user trust. The proposed system utilizes blockchain technol-ogy as an immutable distributed ledger for securely storing vehicle records, ownership history, buyer credentials, and transaction information. Smart contracts are employed to automate critical activities including buyer verification, ownership transfer, payment authorization, and regulatory validation without requiring manual intervention. The decentralized architecture minimizes de-pendency on third-party agencies while improving transparency, reducing processing delays, and enhancing security against data manipulation. Since every transaction is permanently record-ed on the blockchain, both buyers and sellers can independently verify the authenticity of vehicle records before completing a purchase. The proposed framework maintains the same operational workflow and implementation strategy as the reference system while offering improved docu-mentation quality and technical presentation. Experimental observations demonstrate that block-chain-assisted procurement significantly improves transaction efficiency, strengthens security, simplifies ownership transfer, and establishes a reliable digital marketplace for modern automo-tive commerce. The framework represents a scalable solution capable of supporting future intel-ligent transportation systems and smart mobility applications..
Doi-[https://doi.org/10.5281/zenodo.21735050]
An Intelligent Hybrid Artificial Intelligence Framework for Accurate Cyber Threat Detection Using Machine Learning and Deep Learning Techniques
Authors-R.Sankeerthana, Aishwarya Mukkamla
Keyword-Cyber Security, Hybrid Threat Detection, Artificial Intelligence, Machine Learning, Deep Learn-ing, Intrusion Detection System, Network Security, Anomaly Detection, Rule-Based Detection, Cyber Attack Analysis.
Abstract-The rapid growth of digital communication, cloud computing, Internet of Things (IoT), and enterprise networking has significantly increased the complexity and frequency of cyber threats. Conventional intrusion detection mechanisms that rely solely on predefined signatures or static rules often fail to recognize sophisticated attacks, zero-day exploits, and continuously evolving malicious activities. To overcome these limitations, this research proposes a hybrid artificial intelligence-based cyber threat detection framework that integrates machine learning, deep learn-ing, anomaly detection, and rule-based security mechanisms into a unified architecture. The proposed framework continuously analyzes network traffic, user behavior, and system activities to identify both known and unknown cyber attacks with improved accuracy. Machine learning algorithms are employed to discover hidden attack patterns, while deep learning models perform advanced feature learning for complex threat classification. Rule-based verification further vali-dates suspicious events and minimizes false alarms before generating security alerts. The integra-tion of these complementary techniques improves detection performance, enhances adaptability against emerging threats, and reduces false positive rates without changing the original imple-mentation methodology. Experimental observations demonstrate that the hybrid framework provides superior recall, specificity, accuracy, and overall detection efficiency compared with conventional security approaches. The proposed model establishes a scalable, intelligent, and reliable cybersecurity solution capable of protecting modern digital infrastructures against rapidly evolving cyber threats while maintaining the same algorithms and evaluation strategy presented in the original research.
Doi-[https://doi.org/10.5281/zenodo.21735133]
An Intelligent Machine Learning-Based Frame-work for Secure UPI Fraud Detection Using En-semble Classification Models
Authors-S.Venkateswara Rao, Mullamuri yamini
Keyword-UPI Fraud Detection, Machine Learning, Ensemble Learning, XGBoost, Voting Classifier, Stacking Classifier, Digital Payments, Financial Fraud Detection, Flask, SQLite Authentication.
Abstract-The rapid adoption of the Unified Payments Interface (UPI) has transformed digital payment services by enabling instant, convenient, and cashless financial transactions. However, the wide-spread usage of UPI has simultaneously increased the occurrence of fraudulent activities, makingThe proposed framework utilizes Logistic for transaction classification. Furthermore, Voting and Stacking ensemble classifiers are employed to enhance prediction performance by combining the strengths of multiple learning models. The dataset undergoes extensive prepro-cessing, including missing value treatment, feature selection, normalization, and exploratory data analysis to improve learning efficiency. Aenables secure real-time fraud prediction for authorized users. Experimental evaluation demonstrates that ensemble learning significantly improves fraud detection performance, with the Stacking Classifier achieving the highest prediction accuracy while maintaining the same implementation strategy and evaluation methodology as the original study. The proposed framework provides a scalable, intelligent, and reliable solution for protect-ing UPI transactions against fraudulent financial activities while strengthening
Doi-[https://doi.org/10.5281/zenodo.21735199]
An Intelligent Machine Learning Framework for Climate Change Sentiment Analysis Using Twitter Data and Support Vector Machine
Authors-P.Navya, M.Satya Pranitha
Keyword-Climate Change, Sentiment Analysis, Twitter Data, Natural Language Processing, Support Vector Machine, Machine Learning, Text Classification, Social Media Analytics, Opinion Min-ing, Environmental Intelligence.
Abstract-The increasing use of social media platforms has generated vast amounts of public opinion relat-ed to global environmental issues, particularly climate change. Twitter has emerged as one of the most influential platforms where individuals, organizations, and policymakers actively express their views regarding climate-related events and policies. Analyzing these opinions provides valuable insights into public awareness, environmental concerns, and societal attitudes toward climate change. This research presents a machine learning-based sentiment analysis framework that employs Natural Language Processing (NLP) techniques and the Support Vector Machine (SVM) algorithm to classify climate change-related tweets into multiple sentiment categories. A publicly available Twitter dataset obtained from Kaggle is utilized to train and evaluate the pro-posed classification model. Comprehensive preprocessing operations including tokenization, stop-word removal, stemming, and text vectorization are performed to transform unstructured textual data into a machine-readable representation. The trained SVM classifier effectively cap-tures linguistic patterns and semantic relationships present within climate-related discussions. Experimental evaluation demonstrates that the proposed framework achieves reliable sentiment classification performance while maintaining the same implementation methodology and evalua-tion process as the original research. The developed model provides an efficient approach for understanding global public opinion regarding climate change and offers useful information that can support environmental policy development, climate awareness campaigns, and future social media analytics.
Doi-[https://doi.org/10.5281/zenodo.21735215]
An Intelligent Hybrid Deep Learning Framework for Automated Parcel Damage Detection Using Computer Vision and CNN–SVM Classification
Authors-CH.Srinivas Reddy, Muthu Preethi
Keyword-Parcel Damage Classification, Computer Vision, Deep Learning, Convolutional Neural Net-work, Support Vector Machine, Shipment Quality Assessment, Logistics Automation, Image Classification, Damage Detection, Artificial Intelligence.
Abstract-The rapid growth of e-commerce and global logistics has significantly increased the demand for reliable shipment quality inspection systems. Manual parcel inspection methods are often time-consuming, inconsistent, and unsuitable for large-scale logistics operations, leading to increased operational costs and customer dissatisfaction. This research proposes an intelligent computer vision framework that automatically detects and classifies parcel damage using a hybrid deep learning model combining Convolutional Neural Networks (CNN) and Support Vector Ma-chines (SVM). The CNN component is employed to learn high-level visual representations from parcel images, while the SVM classifier performs accurate categorization of damage severity based on the extracted feature vectors. The proposed system identifies multiple parcel conditions, including undamaged parcels, minor damage, moderate damage, and severe damage. A compre-hensive image preprocessing pipeline involving resizing, normalization, and data augmentation improves the robustness of the learning process and enhances model generalization. The devel-oped hybrid framework is trained and evaluated using a large parcel image dataset containing multiple categories of shipment damage. Experimental evaluation demonstrates that the proposed CNN–SVM architecture achieves an overall classification accuracy of 98.8%, indicating its capability for reliable and automated parcel quality assessment. The proposed framework offers a scalable, intelligent, and practical solution for logistics companies, courier services, and e-commerce platforms by minimizing manual inspection efforts, improving shipment quality con-trol, reducing financial losses, and enhancing customer satisfaction while preserving the same implementation methodology and evaluation strategy as the original study.
Doi-[https://doi.org/10.5281/zenodo.21735266]
An Intelligent Machine Learning Framework for Accurate Uber Ride Fare Prediction Using Gradi-ent Boosting Regression
Authors-P.Meghana Sri, Nagaram Jahnavi
Keyword-Uber Ride Prediction, Fare Estimation, Machine Learning, Gradient Boosting Regressor, Ran-dom Forest Regression, Linear Regression, Demand Forecasting, Ride-Hailing Services, Flask, Predictive Analytics.
Abstract-The rapid expansion of app-based transportation services has increased the need for accurate ride fare prediction to improve customer satisfaction and operational efficiency. Reliable fare estima-tion enables passengers to plan their travel expenses while allowing ride-hailing companies to optimize pricing strategies and resource allocation. Conventional fare estimation methods often fail to capture the complex relationships among trip distance, travel time, traffic conditions, and temporal factors, resulting in inconsistent predictions. This research presents a machine learning-based framework for Uber ride fare prediction using historical trip information and regression algorithms. The proposed framework employs comprehensive data preprocessing, feature engi-neering, normalization, and model optimization to enhance prediction performance. Multiple regression algorithms, including Linear Regression, Random Forest Regression, and Gradient Boosting Regressor (GBR), are developed and compared to identify the most effective predictive model. Experimental evaluation demonstrates that the Gradient Boosting Regressor delivers superior prediction performance by effectively learning nonlinear relationships within ride data. The developed model is integrated into a Flask-based web application that provides real-time fare estimation based on user inputs. The proposed framework offers a scalable and intelligent solu-tion for ride fare prediction, improving pricing transparency, operational planning, and customer experience while maintaining the same implementation methodology and experimental evaluation presented in the original study.
Doi-[https://doi.org/10.5281/zenodo.21735287]
An Intelligent Evolving Ensemble Machine Learn-ing Framework for Customer Churn Prediction in Telecommunications
Authors-V.Kranthi Kumar, N sankeerthana
Keyword-Customer Churn Prediction, Machine Learning, Ensemble Learning, Neural Networks, Random Forest, XGBoost, K-Nearest Neighbors, Weighted Average Ensemble, Telecommunications, Predictive Analytics.
Abstract-Customer retention has become one of the most critical challenges faced by telecommunication service providers due to intense market competition and continuously changing customer behav-ior. Accurately identifying customers who are likely to discontinue their subscriptions enables organizations to implement proactive retention strategies and minimize revenue losses. Tradition-al customer churn prediction models frequently struggle to adapt to dynamic customer behavior, resulting in reduced predictive performance over time. This research proposes an intelligent machine learning framework based on an Evolving Ensemble Predictor (EEP) that combines multiple predictive algorithms, including Neural Networks (NN), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and K-Nearest Neighbors (KNN), using a weighted average ensemble strategy. The proposed framework treats customer churn prediction as a continuously evolving problem rather than a static classification task. Historical customer records obtained from the Orange Telecom Churn Dataset are preprocessed through feature encoding, normaliza-tion, and feature selection before model development. Individual classifiers are trained inde-pendently, and their predictions are integrated using weighted ensemble learning to generate the final churn prediction. Experimental evaluation demonstrates that the proposed EEP model achieves improved prediction accuracy, precision, recall, F1-score, and computational efficiency compared with individual machine learning models. The proposed framework provides a scala-ble and adaptive solution for customer churn prediction, assisting telecommunication companies in improving customer retention, reducing operational losses, and supporting intelligent business decision-making while preserving the same implementation methodology and evaluation strategy as the original study.
Doi-[https://doi.org/10.5281/zenodo.21735322]
Telugu Conversational Intelligent Farming Assis-tant with Machine Learning-Based Crop Recommendation and Voice Interaction
Authors-B.Beulah, Sandhya Rani
Keyword-Agricultural Voice Assistant, Crop Recommendation System, Random Forest, Machine Learn-ing, Natural Language Processing, Speech Recognition, Google Text-to-Speech, Telugu Lan-guage Processing, Smart Agriculture, Precision Farming.
Abstract-The agriculture sector is increasingly adopting intelligent technologies to improve farming effi-ciency and decision-making. Farmers in rural regions often encounter difficulties in accessing modern agricultural advisory systems because many available applications rely on text-based interfaces and are developed primarily in English. To overcome these limitations, this research proposes a Telugu-enabled intelligent farming assistant that combines voice interaction with machine learning-based crop recommendation. The proposed framework allows farmers to communicate naturally through speech and receive instant responses in Telugu. Speech recogni-tion techniques convert spoken queries into text, while Natural Language Processing interprets the user's request and retrieves appropriate agricultural information. For crop recommendation, the system employs the Random Forest machine learning algorithm using environmental pa-rameters such as soil pH, rainfall, humidity, and temperature. Google Text-to-Speech technology transforms the generated response into natural Telugu speech, making the application suitable for farmers with limited literacy or technical knowledge. If the requested information is unavailable in the local knowledge repository, the system retrieves relevant agricultural content from online resources to provide comprehensive assistance. The integrated solution improves accessibility to agricultural information, supports informed crop selection, and promotes sustainable farming practices. The proposed system demonstrates that combining machine learning, voice technolo-gy, and regional language support creates a practical digital assistant capable of enhancing agri-cultural productivity while simplifying technology adoption among Telugu-speaking farmers
Doi-[https://doi.org/10.5281/zenodo.21735338]
Machine Learning-Based Intelligent Prediction of Electric Vehicle Battery Health for Enhanced Per-formance and Lifetime Estimation
Authors-CH.Sravan Kumar, Palle Vasantha
Keyword-Electric Vehicles (EV), Battery State of Health, Battery Management System, Machine Learning, XGBoost, LightGBM, Battery Performance Prediction, Lithium-Ion Battery, Predictive Analyt-ics
Abstract-The increasing adoption of electric vehicles (EVs) has created a growing demand for reliable battery health monitoring systems that can improve operational efficiency and extend battery service life. One of the most important performance indicators of a lithium-ion battery is its State of Health (SOH), which reflects the remaining capacity and overall condition of the battery throughout its lifecycle. Accurate SOH estimation enables timely maintenance, minimizes unex-pected failures, and improves the reliability of electric transportation systems. This research presents a machine learning framework for predicting EV battery SOH using two ensemble learning algorithms, namely Extreme Gradient Boosting (XGBoost) and Light Gradient Boost-ing Machine (LightGBM). The proposed framework utilizes battery operating parameters in-cluding voltage, current, temperature, and charging history to develop predictive models capable of learning battery degradation patterns. The collected dataset undergoes preprocessing, feature engineering, and train-test partitioning before model training. Performance evaluation is carried out using statistical measures such as R² Score, Adjusted R², Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE). Experimental findings demonstrate that both algorithms achieve excellent prediction accuracy, with XGBoost showing slightly superior performance compared to LightGBM. The developed system provides an effi-cient solution for real-time battery health assessment, supporting intelligent maintenance schedul-ing, reducing operational costs, and improving battery lifespan in modern electric vehicles.
Doi-[https://doi.org/10.5281/zenodo.21735381]
An Intelligent Machine Learning Framework for Health Insurance Claim Cost Prediction Using Healthcare Risk Factors
Authors-M.Anitha, Patibandla Sushmitha
Keyword-Health Insurance, Insurance Claim Prediction, Machine Learning, Healthcare Analytics, Random Forest Regressor, Gradient Boosting Regressor, Support Vector Regression, Linear Regression, Predictive Analytics, Cost Forecasting, Healthcare Risk Assessment, Data Mining.
Abstract-For insurance companies to maximize premium computation, enhance financial planning, and facilitate efficient risk management, accurate cost projections of health insurance claims are cru-cial. The intricate links between lifestyle factors, health issues, and demographic traits that are frequently difficult for conventional statistical prediction tools to capture might occasionally hinder prediction performance. Using a variety of healthcare-related factors, such as age, gender, body mass index (BMI), smoking habits, number of dependents, genetic illnesses, occupation, and residential area, the intelligent machine learning method presented in this study predicts health insurance claim expenses. To increase forecasting accuracy and decision-making, the suggested approach combines feature engineering, predictive modeling, and meticulous data pretreatment. Standard performance metrics like R-squared (R²) and Mean Absolute Error (MAE) are used to create and assess regression approaches including Linear Regression, Sup-port Vector Regression (SVR), Random Forest Regressor, and Gradient Boosting Regressor. An experimental study found that while Gradient Boosting also produces comparable forecast-ing performance, the Random Forest Regressor offers the highest prediction accuracy by suc-cessfully capturing nonlinear linkages among health-related indicators. Insurance businesses can improve pricing strategies, lower financial uncertainty, anticipate future claim costs more accu-rately, and construct customized insurance policies more easily with the help of the suggested framework. By offering scalable, dependable, and data-driven prediction capabilities appropriate for contemporary health insurance administration systems, the developed method further illus-trates the useful advantages of machine learning for intelligent healthcare analytics.
Doi-[https://doi.org/10.5281/zenodo.21735431]
An Efficient Deep Convolutional Neural Network Framework for Automated Multiclass Classifica-tion of White Blood Cells from Microscopic Blood Smear Images
Authors-P.Shilpa, Pooja Pawar
Keyword-decision trees, computer-aided diagnosis, medical image analysis, leukocyte identification.
Abstract-White blood cell (WBC) subtype identification is essential for the detection of leukemia, infec-tions, immunological deficits, and hematological disorders. When processing a large number of blood smear samples, laboratory specialists' traditional microscopic examination is labor-intensive, time-consuming, and subject to subjective interpretation. An automated multiclass white blood cell categorization system based on Deep Convolutional Neural Networks (CNNs) is presented in this work to overcome these issues. Neutrophils, monocytes, lymphocytes, and eosinophils are the four main leukocyte categories that the suggested approach divides micro-scopic blood smear images into.techniques like scaling, normalization, and dataset splittin g. A deep CNN architecture is used to automatically find discriminative picture features without the need for manually generated feature engineering, while a Decision Tree model functions as a baseline machine learning classifier The suggested CNN model achieves over 97% classification accuracy whereas the Decision Tree classifier achieves roughly 32.2%, demonstrating deep learning's superior ability to extract meaningful visual representations from tiny images, accord-ing to experimental results. By cutting down on analysis time and increasing diagnostic con-sistency,e replacement for computer-aided hematological diagnosis and can greatly improve clinical decision-making for medical professionals.
Doi-[https://doi.org/10.5281/zenodo.21735474]
An Intelligent Healthcare Conversational Assistant Using Gamma LLM V2 with Comparative Evalu-ation of Transformer-Based BERT Models
Authors-T.Thirumalash, P.Mahalaxmi
Keyword-Healthcare Chatbot, Gamma LLM V2, Medical Question Answering, Large Language Models (LLMs), BERT, MedBERT, RoBERTa, Sentence-BERT (SBERT), LangChain, Pinecone Vec-tor Database, Retrieval-Augmented Generation (RAG), Natural Language Processing (NLP), Semantic Search, Artificial Intelligence, Medical Decision Support.
Abstract-The rapid advancement of artificial intelligence has significantly improved the capabilities of intelligent healthcare applications by enabling automated medical consultation and decision sup-port. This research presents a medical conversational assistant powered by Gamma LLM V2, designed to provide accurate, context-aware responses related to symptoms, medications, diseas-es, and dietary recommendations. The proposed framework employs Retrieval-Augmented Generation (RAG) by combining semantic vector retrieval with a large language model to im-prove the reliability of generated responses. Medical knowledge is extracted from a comprehen-sive medical reference document, segmented into meaningful text chunks, transformed into vector embeddings, and stored within the Pinecone vector database for efficient semantic retriev-al. LangChain is utilized to orchestrate document retrieval and prompt generation, while Flask and Streamlit provide backend services and an interactive user interface. To evaluate the effec-tiveness of the proposed chatbot, Gamma LLM V2 is compared with widely adopted transform-er-based language models including BERT, RoBERTa, MedBERT, and Sentence-BERT (SBERT). Experimental analysis demonstrates that Gamma LLM V2 consistently produces more accurate, contextually relevant, and coherent responses while maintaining superior perfor-mance across medical question-answering tasks. The comparative evaluation shows performance scores of 0.95 for MedBERT, 0.92 for SBERT, 0.86 for BERT, and 0.77 for RoBERTa, con-firming the effectiveness of the proposed approach. The developed system offers a scalable and practical AI-assisted healthcare solution that improves access to preliminary medical guidance while reducing response time and enhancing user experience.
Doi-[https://doi.org/10.5281/zenodo.21735650]
Machine Learning-Based Predictive Framework for Early Identification of Mental Health Disorders
Authors-Pitla Shravani, K.Rajkumar
Keyword-Mental Health Prediction, Machine Learning, Supervised Learning, Logistic Regression, Sup-port Vector Machine, Psychological Disorder Detection, Healthcare Analytics, Questionnaire-Based Screening.
Abstract-Mental health disorders have become a significant public health concern due to increasing aca-demic, professional, and social pressures experienced by individuals worldwide. Early identifi-cation of psychological conditions can substantially improve treatment outcomes and reduce long-term complications. This research presents an intelligent machine learning framework for automated mental health assessment using structured self-report questionnaires. The proposed system utilizes two questionnaire modules to identify general mental health conditions and clas-sify five common psychological disorders, namely Bipolar Disorder, Anxiety Disorder, Depres-sion, Eating Disorder, and Sleep Disorder. Supervised machine learning algorithms including Logistic Regression, Decision Tree, Support Vector Machine (Linear and RBF kernels), and Naïve Bayes are employed to analyze questionnaire responses and generate predictive outcomes. Logistic Regression is adopted for initial mental health screening, while Support Vector Machine with a linear kernel demonstrates superior performance for multiclass disorder identification. The experimental evaluation conducted on 1,253 valid questionnaire responses confirms the effec-tiveness of the proposed framework, achieving high classification accuracy while maintaining a simple and cost-effective implementation. The developed model can serve as an efficient prelimi-nary screening tool to support healthcare professionals, educational institutions, and individuals in recognizing potential mental health risks at an early stage.
Doi-[https://doi.org/10.5281/zenodo.21735677]
A Comprehensive Study of Intelligent Document Image Layout Analysis Using Traditional Image Processing and Deep Learning Techniques
Authors-P.Shilpa, Police Patel Radha
Keyword-Document Layout Analysis, OCR, Deep Learning, Image Segmentation, Connected Component Analysis, CNN, Graph Neural Networks, Text Detection, Skew Correction, Document Pro-cessing.
Abstract-Document image layout analysis has become an essential preprocessing stage for modern Opti-cal Character Recognition (OCR) systems because the accuracy of text extraction depends heavi-ly on preserving the structural organization of document pages. Documents such as newspapers, books, magazines, invoices, historical manuscripts, and research articles usually contain complex layouts consisting of paragraphs, images, tables, figures, mathematical expressions, and multiple font styles. Conventional OCR systems often fail to maintain these structures, leading to incor-rect reading sequences and reduced recognition performance. This paper presents a detailed survey of existing document layout analysis approaches developed using both traditional image processing techniques and modern deep learning methods. Various segmentation, skew detec-tion, text and non-text separation, connected component analysis, projection profile methods, graph neural networks, convolutional neural networks, and hybrid learning models are reviewed and compared based on datasets and evaluation metrics reported in the literature. The study also examines preprocessing techniques that improve document quality before layout segmentation. A comparative analysis highlights the advantages and limitations of existing approaches while identifying research gaps for multilingual and highly complex document layouts. The survey concludes that integrating classical image processing with deep learning provides superior per-formance compared to standalone methods and offers a promising direction for future document understanding systems.
Doi-[https://doi.org/10.5281/zenodo.21735696]
An Intelligent Machine Learning Framework for Public Transport Passenger Demand Forecasting Using Time Series and Regression Models
Authors-M.Swathi, Srishti Kulkarni
Keyword-Passenger Demand Prediction, Public Transportation, Machine Learning, Prophet Model, ARIMA, Random Forest, Decision Tree, Linear Regression, Time Series Forecasting, Smart Transportation.
Abstract-Accurate prediction of passenger demand is essential for improving the efficiency, reliability, and resource management of urban public transportation systems. Reliable forecasting enables transport authorities to optimize vehicle scheduling, reduce passenger waiting times, and enhance the overall quality of service. This study presents a comparative machine learning framework for forecasting passenger demand across stations of Lima Metro Line 1 using five predictive algo-rithms: Prophet, Linear Regression, Random Forest, Decision Tree, and AutoRegressive Inte-grated Moving Average (ARIMA). Historical passenger records collected from the official OSITRAN transportation database are utilized to train and evaluate the forecasting models. Prior to model development, the dataset undergoes comprehensive preprocessing, including data inte-gration, cleaning, datetime transformation, and passenger count aggregation to ensure consisten-cy and reliability. The trained models are assessed using Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R²) to determine their predic-tive capability. Experimental analysis demonstrates that the Prophet model consistently achieves the highest forecasting accuracy, producing the lowest prediction errors and the highest R² value among all evaluated models. Although Decision Tree and Linear Regression require considera-bly less computational time, they exhibit lower predictive performance compared with Prophet. The findings indicate that Prophet effectively captures temporal demand patterns and seasonal variations present in passenger transportation data, making it a practical and reliable solution for intelligent public transport planning, demand forecasting, and operational decision support in modern urban transit systems.
Doi-[https://doi.org/10.5281/zenodo.21735729]
An Intelligent Machine Learning Framework for Predictive Maintenance in Smart Manufacturing Systems
Authors-P.Ramakrishna, T.Sai Nikitha
Keyword-Predictive Maintenance, Machine Learning, Industry 4.0, Random Forest, XGBoost, Support Vector Machine, Equipment Failure Prediction, Industrial Automation, Data Analytics, Condi-tion Monitoring.
Abstract-The growing for intelligent maintenance strategies capable of preventing unexpected equipment failures. Predictive maintenance has emerged as an effective solution by utilizing historical op-erational data and machine learning techniques to estimate machine health before critical failures occur. This research presents a comprehensive The proposed framework employs Random Forest, Extreme Gradient Boosting (XGBoost), to analyze operational sensor measurements and identify possible machine failures. collected from industrial machines. Data preprocessing in-volves label encoding, feature scaling, train-test splitting, and robust normalization before model development. Five-feffectively minimizes unexpected downtime, improves maintenance schedul-ing, lowers operational costs, and enhances manufacturing productivity. These findings indicate that supervised machine learning provides a reliable and scalable solution for intelligent mainte-nance planning across modern industrial environments.
Doi-[https://doi.org/10.5281/zenodo.21735759]
Deep Learning-Based Human Pose Anomaly Recognition Using Stable Diffusion Generated Images and EfficientNetV2 Classification
Authors-S.Akhila, T. kavya
Keyword-Human Pose Anomaly Detection, Stable Diffusion, EfficientNetV2, Deep Learning, Image Classification, Industrial Safety, Artificial Intelligence, Worker Monitoring, Computer Vision, Synthetic Dataset.
Abstract-Human pose anomaly recognition plays a significant role in improving workplace safety, intelli-gent surveillance, healthcare monitoring, and human–robot collaboration. Conventional ap-proaches generally rely on pose estimation algorithms to extract skeletal keypoints before per-forming anomaly classification, making the overall pipeline computationally complex and highly dependent on pose estimation accuracy. This research introduces a simplified image-based anomaly detection framework that directly classifies worker poses without employing any inter-mediate pose estimation module. A synthetic dataset containing both normal and abnormal in-dustrial worker poses is generated using the Stable Diffusion image generation model, allowing the creation of a large and consistent training dataset with well-defined anomaly categories. The generated images are subsequently used to train EfficientNetV2 deep convolutional neural net-work variants for binary pose classification. The proposed framework reduces processing com-plexity while maintaining excellent recognition capability. Experimental evaluation demonstrates that the EfficientNetV2-M architecture provides the highest classification performance, achieving an accuracy of 95.66%, outperforming the remaining EfficientNetV2 variants. The findings indicate that direct image classification is capable of learning discriminative pose representations without requiring explicit skeletal information. The proposed framework provides an efficient, scalable, and practical solution for industrial safety monitoring and intelligent worker surveil-lance applications.
Doi-[https://doi.org/10.5281/zenodo.21735776]
An Intelligent Machine Learning Framework for Predicting Customer Purchase Decisions Using Classification and Regression Techniques in E-Commerce
Authors-S.Srinivas, Thumu Pradeepthi
Keyword-Customer Purchase Prediction, Machine Learning, Random Forest, Decision Tree, Logistic Regression, Feature Engineering, Classification, E-Commerce Analytics, Consumer Behavior, Predictive Modeling
Abstract-The rapid expansion of e-commerce platforms has generated enormous volumes of consumer interaction data, creating new opportunities for understanding purchasing behavior through intelligent data analytics. Accurate prediction of customer purchase decisions enables online retailers to improve personalized recommendations, optimize promotional campaigns, and en-hance customer satisfaction. This study proposes a machine learning framework for predicting whether a customer will purchase a product after browsing it on an e-commerce platform. The research utilizes a large-scale transaction dataset obtained from JD.com containing customer information, product characteristics, browsing history, pricing details, and promotional activities. Before model construction, comprehensive data preprocessing and feature engineering tech-niques are performed to improve data quality and create informative predictive variables. The proposed framework evaluates three supervised machine learning algorithms, namely Decision Tree, Random Forest, and Logistic Regression, for binary purchase classification. Hyperparame-ter optimization is conducted using the Random Search strategy to maximize prediction perfor-mance while reducing overfitting. Model effectiveness is evaluated using Accuracy, Precision, Recall, and Area Under the ROC Curve (AUC). Experimental results demonstrate that the opti-mized Random Forest classifier achieves the highest prediction performance with a testing accu-racy of 0.999871 and an AUC of 0.9998, outperforming the remaining models. Feature im-portance analysis further indicates that coupon discount level and quantity discount level con-tribute most significantly to customer purchasing decisions. The proposed framework offers a reliable and computationally efficient solution for customer behavior prediction and provides valuable support for personalized recommendation systems, pricing optimization, inventory planning, and intelligent marketing strategies in modern e-commerce platforms.
Doi-[https://doi.org/10.5281/zenodo.21735797]
An Intelligent Deep Learning Framework for Au-tomated Cattle Breed Recognition Using Image-Based Feature Analysis
Authors-S.Venkateswara Rao, Udari Santhoshma
Keyword-Deep Learning, Cattle Breed Classification, Convolutional Neural Network, DenseNet201, MobileNetV2, InceptionV3, Xception, Image Processing, Transfer Learning, Livestock Moni-toring, Precision Agriculture.
Abstract-The identification of cattle breeds is an important task in precision livestock farming because it assists farmers in maintaining breed quality, improving breeding strategies, and enhancing over-all farm productivity. Conventional breed identification methods generally rely on manual obser-vation, making the process labor-intensive, time-consuming, and susceptible to human error. Recent advances in artificial intelligence have enabled image-based automated systems capable of performing accurate breed classification with minimal human intervention. This study presents an automated cattle breed recognition framework based on deep convolutional neural networks. The proposed approach employs image preprocessing techniques including resizing, normaliza-tion, background enhancement, and augmentation to improve image quality before classification. Four well-established transfer learning architectures—DenseNet201, MobileNetV2, Incep-tionV3, and Xception—are utilized to extract discriminative visual features from cattle images belonging to multiple breeds. The trained models are evaluated using standard performance metrics such as accuracy, precision, recall, F1-score, confusion matrix, and ROC analysis. Ex-perimental evaluation demonstrates that the Xception architecture provides superior classification performance among the evaluated models while maintaining strong generalization capability. The proposed framework offers an efficient, scalable, and reliable solution for automatic cattle breed identification, thereby supporting intelligent livestock management, genetic conservation, and sustainable agricultural practices.
Doi-[https://doi.org/10.5281/zenodo.21735820]
Deep Learning-Based Intelligent Monument Recognition and Augmented Reality Visualization for Interactive Cultural Heritage Exploration
Authors-S.Venkateswara Rao, Udutha Bhanu Sri
Keyword-Deep Learning, Monument Recognition, Cultural Heritage, VGG16, Convolutional Neural Network, Transfer Learning, Image Classification, Augmented Reality, Android Application, Computer Vision
Abstract-The preservation and promotion of historical monuments have become increasingly important as digital technologies continue to reshape cultural education and tourism. Conventional monument recognition techniques generally rely on manual searches or static information systems, making it difficult for visitors to obtain accurate and engaging historical information during site visits. To overcome these limitations, this research introduces an intelligent monument recognition frame-work that combines deep learning with augmented reality to provide an interactive cultural herit-age experience. The proposed system employs the VGG16 convolutional neural network as the primary feature extraction model for recognizing monuments captured through mobile cameras. A lightweight convolutional neural network is also implemented to compare classification per-formance under identical experimental conditions. Before classification, images undergo prepro-cessing operations including resizing, normalization, grayscale conversion, and feature en-hancement. The recognized monument is linked with historical descriptions, geographical loca-tion, and three-dimensional augmented reality visualization, allowing users to explore monu-ments through an immersive digital interface. The complete framework is deployed as an An-droid application, enabling real-time monument identification and visualization using mobile devices. Experimental evaluation demonstrates that the transfer learning capability of VGG16 produces superior classification performance compared with the conventional CNN architecture. The VGG16 model achieves an overall testing accuracy of 98.01%, whereas the standard CNN records 95.34% accuracy using the same dataset. The integration of augmented reality further improves user engagement by presenting historical information within the surrounding environ-ment instead of traditional text-based interfaces. The proposed framework offers a practical solution for digital heritage preservation, smart tourism, and educational applications while main-taining high recognition accuracy under different viewing conditions.
Doi-[https://doi.org/10.5281/zenodo.21735846]
An Intelligent Cyber Threat Intelligence Frame-work for Network Indicator Analysis and Proac-tive Threat Detection Using MITRE ATT&CK
Authors-P.Premchand Goud, Valli Mounika
Keyword-Cyber Threat Intelligence, Cybersecurity, MITRE ATT&CK, Indicators of Compromise, Threat Detection, Network Security, STIX, TAXII, Intrusion Detection System, Sigma Rules, Vulner-ability Management, YARA, SIEM, Threat Hunting
Abstract-The increasing dependence on digital infrastructure has significantly expanded the cyber threat landscape, making organizations more vulnerable to sophisticated cyber-attacks. Traditional security mechanisms often rely on reactive defense strategies that struggle to identify emerging threats before they compromise critical assets. To address these challenges, this research propos-es an intelligent Cyber Threat Intelligence (CTI) framework that utilizes technical threat indica-tors to strengthen proactive cyber defense. The proposed framework focuses on the collection, processing, correlation, and consumption of network-based indicators such as malicious IP addresses, suspicious domains, intrusion detection signatures, and vulnerability intelligence. Threat information is organized using standardized CTI formats and mapped to the MITRE ATT&CK framework to improve the understanding of attacker behavior and support effective defensive decision-making. The framework further integrates network intrusion detection sys-tems, vulnerability repositories, Sigma rules, YARA signatures, and threat intelligence feeds to enable continuous monitoring and rapid detection of malicious activities. By combining struc-tured intelligence with real-time security monitoring, the proposed solution enhances threat visi-bility, reduces response time, and supports proactive mitigation against advanced cyber threats. The framework also assists security analysts in prioritizing vulnerabilities, identifying attack patterns, and strengthening organizational cyber resilience through intelligence-driven security operations.
Doi-[https://doi.org/10.5281/zenodo.21735863]
Interpretable Machine Learning for Predicting Climate Change Effect on Agricultural Land Suitability in Eurasia
Authors-
Keyword-Climate Change, Agricultural Land Suitability, Interpretable Machine Learning, Explainable AI, SHAP, Random Forest, Gradient Boosting, Eurasia, Spatial Analysisltural policy and climate adaptation strategies.
Abstract-Temperature and precipitation patterns are changing quickly due to climate change, which pre-sents significant obstacles to the sustainability of agriculture throughout Eurasia. Effective long-term planning requires accurate forecasting of changes in the suitability of agricultural land. High precision and interpretability are frequently lacking in conventional statistical models. An inter-pretable machine learning method for evaluating the effects of climate change on the usability of agricultural land is presented in this study. The input features include a variety of climate indica-tors, soil characteristics, and land-use factors. Prediction is the goal of machine learning models like Random Forest and Gradient Boosting. Standard accuracy and error measures are used to assess the model's performance. Interpretability strategies like SHAP are used to improve trans-parency. The relative significance of environmental and climatic elements is revealed by these techniques. To find geographical differences, spatial analysis is done.
Doi-[https://doi.org/10.5281/zenodo.21735896]
An Intelligent Machine Learning Framework for Accurate Uber Ride Fare Prediction Using Gradient Boosting Regression
Authors-Dr. M. A. Azeem, Yatakarla Sandhyarani
Keyword-Uber Fare Prediction, Machine Learning, Gradient Boosting Regressor, Random Forest, Linear Regression, Intelligent Transportation Systems, Ride-Hailing Services, Regression Analysis, Predictive Analytics, Flask Web Application.
Abstract-The rapid expansion of ride-hailing platforms has transformed urban transportation by offering convenient, flexible, and on-demand mobility services. As the number of daily ride requests continues to increase, accurately estimating ride fares has become essential for improving opera-tional efficiency, enhancing customer satisfaction, and supporting intelligent transportation man-agement. Reliable fare prediction enables passengers to estimate travel costs before booking while assisting service providers in optimizing pricing strategies and resource allocation. This research presents a machine learning-based framework for predicting Uber ride fares using historical trip information and environmental attributes. The proposed framework utilizes public-ly available Uber trip records collected from the NYC Open Data repository, containing infor-mation such as travel distance, journey duration, time of day, and fare values. Before model development, the dataset undergoes comprehensive preprocessing involving missing value han-dling, feature engineering, normalization, and data partitioning to improve data quality and model performance. Three regression algorithms, namely Linear Regression, Random Forest Regres-sor, and Gradient Boosting Regressor (GBR), are implemented and evaluated under identical experimental conditions. Hyperparameter optimization is performed for the Gradient Boosting model by adjusting the number of estimators, learning rate, and tree depth to maximize predictive accuracy while preventing overfitting. Model performance is evaluated using the Coefficient of Determination (R²) and Mean Squared Error (MSE). Experimental analysis demonstrates that the Gradient Boosting Regressor achieves the highest predictive performance, producing a train-ing accuracy of 99.99% and a testing accuracy of 91.79%, outperforming both Random Forest and Linear Regression models. Furthermore, feature analysis identifies travel distance and time of day as the most influential variables affecting fare estimation. The optimized prediction model is integrated into a Flask-based web application that enables users to obtain real-time fare esti-mates through an interactive interface. The proposed framework provides a scalable, accurate, and practical solution for intelligent ride fare prediction and supports the development of efficient transportation management systems.
Doi-[https://doi.org/10.5281/zenodo.21735912]
An Intelligent XGBoost-Based Machine Learning Framework for Accurate Life Expectancy Prediction Across Developed and Developing Countries
Authors-S.Gouthami, Yeguri Supriya
Keyword-Life Expectancy Prediction, XGBoost, Machine Learning, Feature Selection, M5P Algorithm, RandomizedSearchCV, Healthcare Analytics, Predictive Modeling, Regression Analysis, Public Health.
Abstract-Life expectancy is one of the most important indicators used to evaluate the overall health status, healthcare quality, and socio-economic development of a country. Accurate prediction of life expectancy enables governments and healthcare organizations to formulate effective public health policies and allocate medical resources more efficiently. In recent years, machine learning tech-niques have demonstrated superior capability in modeling complex healthcare datasets compared with conventional statistical approaches. This study presents a robust life expectancy prediction framework based on the Extreme Gradient Boosting (XGBoost) algorithm for estimating life expectancy in both developed and developing countries. Initially, the collected dataset undergoes comprehensive preprocessing, including missing value imputation, normalization, and categori-cal encoding to improve data quality. The M5P decision tree algorithm is then employed to iden-tify the most informative attributes influencing life expectancy. Subsequently, the processed data are divided into training and testing subsets, and the XGBoost regression model is optimized using RandomizedSearchCV to determine the most suitable hyperparameter configuration. Mod-el performance is evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R²). Experimental analysis indicates that the optimized XGBoost model achieves excellent predictive performance with a testing R² score of 0.97, MAE of 1.03, and MSE of 2.45, outperforming several conven-tional machine learning approaches. Feature importance analysis further reveals that HIV/AIDS prevalence, adult mortality, and income composition are among the strongest determinants affect-ing life expectancy. The proposed framework provides an efficient and reliable decision-support tool for healthcare planners, policymakers, and researchers working toward improving global health outcomes.
Doi-[https://doi.org/10.5281/zenodo.21703353]
Investigating Barriers to Effective Chemistry Teaching in Rural Areas: A Case Study Analysis
Authors-Mwale Phillip, Dr Phiri John
Keyword-Chemistry Education, Rural Areas, Education Barriers, Socio-economic Factors, STEM Educa-tion, Case Study, And Purpose Sampling
Abstract-This study explored the barriers to effective chemistry teaching and learning through a qualitative case study of selected rural secondary schools. Chemistry, as a practical and conceptually de-manding subject, requires adequate laboratory facilities, instructional resources, supportive so-cio-economic environments, and accessible school infrastructure to ensure meaningful learning. However, rural schools often experience contextual challenges that compromise the delivery of quality chemistry education. This study sought to examine how resource limitations, socio-economic factors, geographical isolation, and infrastructure challenges affect chemistry teaching and learning in the selected rural school context. A qualitative case study design was employed to obtain in-depth insights into the lived experiences of teachers and learners. The study involved a purposive sample of forty (40) participants, comprising ten (10) chemistry teachers and thirty (30) Grade 12 learners. Data were collected using semi-structured interview guides for both teachers and learners. The data were analysed using thematic analysis to identify recurring pat-terns, themes, and relationships related to the research objectives. The findings revealed that resource limitations constituted the most significant barrier to effective chemistry teaching. The absence of functional laboratories, inadequate chemicals and apparatus, and limited access to up-to-date textbooks resulted in reduced practical experimentation and an overreliance on teacher-centred instructional methods. Socio-economic challenges, including poverty, limited parental support, and learner involvement in household responsibilities, negatively affected attendance, concentration, and time allocated for homework and revision. Geographical isolation constrained teachers’ access to professional development and delayed the supply of laboratory materials, while also limiting learners’ exposure to academic enrichment opportunities. Infrastructure defi-ciencies, such as unreliable electricity and water supply, overcrowded classrooms, and poor laboratory conditions, further restricted the safe and effective implementation of practical chemis-try lessons. The study concluded that these barriers operate cumulatively within the rural school context, significantly undermining learners’ conceptual understanding, practical competence, motivation, and academic performance in chemistry. The case study highlights the urgent need for targeted policy interventions and resource allocation strategies aimed at strengthening rural science education. The study recommends increased investment in laboratory infrastructure, improved distribution of teaching materials, enhanced professional development programs for rural chemistry teachers, and community-based initiatives to address socio-economic constraints affecting learners. Addressing these barriers is critical to promoting equitable access to quality chemistry education in rural secondary schools.
Doi-[https://doi.org/10.5281/zenodo.21737630]
Examining the Effectiveness of Digital Learning Platforms in Early Literacy Development: A Case Study of Three Selected Primary Schools in Katete District, Eastern Zambia
Authors-Phiri Clement, Dr. Phiri John
Keyword-Digital learning platforms, Early literacy development, Primary education, Literacy skills, Educational technology, Digital education
Abstract-Early literacy development is a critical foundation for learners’ academic success and lifelong learning. Despite increased efforts to improve early-grade reading outcomes in Zambia, many learners in lower primary grades continue to demonstrate low levels of literacy proficiency. In response to these challenges, digital learning platforms have increasingly been introduced in primary schools as supplementary instructional tools aimed at enhancing early literacy skills. However, there is limited empirical evidence on the effectiveness of such platforms, particularly in rural districts. This study examined the effectiveness of digital learning platforms in early literacy development through a case study of three selected primary schools in Katete District, Eastern Zambia. The study adopted a mixed-methods case study design, combining quantitative and qualitative approaches to provide a comprehensive evaluation of digital learning platform use. Quantitative data were collected using baseline and post-test early literacy assessments administered to Grade 1 and Grade 2 learners, while qualitative data were obtained through teachers’ questionnaires, interviews with school administrators, and classroom observations. The findings revealed that learners exposed to digital learning platforms demonstrated measurable improvement in key early literacy skills, particularly letter recognition, phonemic awareness, and word decoding. Improvements were also observed in oral reading fluency and reading comprehension, although gains in these areas were relatively modest. Teachers generally perceived digital learning platforms as effective in enhancing learner engagement and motivation, especially through audio-visual and interactive features. However, the effectiveness of the platforms was influenced by contextual and implementation factors, including availability of digital devices, reliability of electricity supply, teacher competence, and consistency of platform use. The study concludes that digital learning platforms can play a positive role in supporting early literacy development in rural primary schools when integrated into regular classroom instruction and supported by adequate infrastructure and teacher training. The study recommends increased investment in digital infrastructure, continuous professional development 1 for teachers, and structured integration of digital learning platforms into early literacy programmes. The findings contribute context-specific empirical evidence to inform educational policy, practice, and future research on digital learning and early literacy development in Zambia.
Doi-[https://doi.org/10.5281/zenodo.21773432]
The Impact of Free Education Policy on the Infrastructure and Learning Materials at Primary School Level: A Case of Four Selected Schools of Mambwe District, Eastern Province, Zambia
Authors-Sakala Weston Mmembe, Dr Phiri John
Keyword-Impact, education policy, learning, infrastructure, learning materials, primary schools.
Abstract-The study sought to establish the impact of free education policy on infrastructure and learning materials at primary school level in selected schools of Mambwe District. The objectives of the study were: to establish the state of the infrastructure and learning materials in the context of the free-education policy in selected primary schools of Mambwe District; to investigate the impact of the current state of the infrastructure and learning materials on the teaching and learning process in selected primary schools of Mambwe District; to establish the role of school managers in enhancing proper infrastructure and learning materials in selected primary schools of Mambwe District; and to recommend possible measures to improve the infrastructure and leaning materials in selected primary schools of Mambwe District. Descriptive survey design namely cross-sectional was used and employed mixed methods but with a greater focus on qualitative research. The target group was drawn from the selected schools of Mambwe District and District Education standards officers giving a total number of 69. 3 Standards officers and the 8 head teachers were sampled using the non-probability procedure engaging purposive sampling technique while the 16 teachers and the 40 pupils were sampled using simple random sampling method. The instruments for collecting data were: interviews schedules for standards officers and head teachers, questionnaires for teachers, Focus Group Discussions (FGD) for pupils and observations. Data collected from questionnaires was analyzed using the Statistical Package for Social Sciences (SPSS) while data collected from interviews and FGD was analyzed using qualitative thematic analysis. After the FPE policy, the physical infrastructure and learning materials increased thus fairly adequate but could not cater for the high enrolments numbers of pupils causing overcrowding and poor teaching and learning. The significance of the study was to benefit the stakeholders in knowing the constraints of inadequate physical infrastructures and learning materials that needed their contribution. Policy makers would identify strategies and re-define the policy framework on the provision to curb the congestion, add 1 knowledge in the area and identify the gaps that needed further research. The findings indicated inadequate and poor condition of infrastructure and learning materials hindering pupils’ space and access hence poor quality education. Based on the findings, the study made recommendations directed to the infrastructure and learning materials which were grounded on the improvements in the free education policy
Doi-[https://doi.org/10.5281/zenodo.21774239]
Assessing the Effectiveness of Competency-Based Curriculum (CBC) Implementation in Biology Education in Selected Secondary Schools in Kafue District
Authors-Likando Likando
Keyword-Competency-Based Curriculum (CBC), Biology Education, Curriculum Implemen-tation, Secondary Schools, Kafue District, Teaching and Learning
Abstract-This study assessed the effectiveness of the Competency-Based Curriculum (CBC) implementation in Biology education in selected secondary schools in Kafue District. The main objective of the study was to evaluate how effectively the CBC has been implemented in teach-ing and learning Biology, focusing on teacher preparedness, availability of teaching and learning materials, teaching methods, and challenges faced during implementation. A descriptive survey research design was used in this study. The target population included head teachers, Biology teachers, and pupils from selected secondary schools. A sample was selected using purposive and random sampling techniques. Data were collected using questionnaires, interviews, and ob-servation checklists. The findings of the study revealed that while teachers had a general under-standing of the CBC, many faced challenges such as inadequate training, lack of teaching re-sources, large class sizes, and limited laboratory facilities. The study also found that although learner-centered teaching methods were encouraged under CBC, their application was incon-sistent due to these challenges. The study concluded that the implementation of CBC in Biology education in Kafue District is moderately effective but requires significant improvement to achieve its intended goals. The study recommended increased teacher training, improved provi-sion of teaching materials, and enhanced support from educational authorities to strengthen CBC implementation.
Doi-[https://doi.org/10.5281/zenodo.21788430]
Teachers’ Role in Promoting Caring Classroom and Creating Inclusive Learning Environment in Elementary Schools
Authors-Sumia Rasool, Dr. Naushad Husain
Keyword-Caring classroom, Inclusive learning environment, Elementary schools, Teachers’ role, Barriers, Relevant instructional strategies, Caring pedagogy and inclusive practices, Quality education, Diverse learner, Holistic child development.
Abstract-In the present educational scenario, elementary schools are expected to function not only as centers of academic learning but also spaces for emotional safety, inclusion, empathy, and holis-tic child development. The growing diversity among children increases the necessity of creating caring and inclusive learning environments that promote both academic and socio-emotional growth of the diverse learner. This paper explores the central, multidimensional role of teachers in nurturing caring classrooms and fostering equitable learning opportunities when the existence of the diverse learners is inevitable. In agreement with the existing literature and core theoretical foundations, the paper examines how teachers can successfully integrate emotional care and professional knowledge to nurture belongingness, mutual respect, and active participation of the children. While inclusive education allows all students to learn under a single roof without dis-crimination, at the same time teachers face significant systemic challenges in creating inclusive environment, containing emotional burnout, large class sizes, resource scarcity, and a lack of professional training. To address these barriers, this paper highlights the value of adopting rele-vant instructional strategies and proposes practical methods for strengthening teacher prepared-ness and school culture for creating inclusive environment in the school. Lastly, this paper ar-gues that caring pedagogy and inclusive practices are interconnected, essential dimensions of quality education necessary for achieving equity, quality, diversity, accessibility, and holistic development in elementary education.
Doi-[https://doi.org/10.5281/zenodo.21898831]
Social Competence of Adolescents in Relation to Their Emotional Maturity
Authors-Simran, Assistant Professor, MR.Jagbir Grewal
Keyword-Social Competence, Emotional Maturity, Adolescents, Gender Difference, Descriptive Survey.
Abstract-Adolescence is a stage marked by rapid emotional and social change, and the manner in which a young person regulates emotion is closely bound up with the manner in which the same young person relates to others. The present paper reports the findings of a study undertaken to examine the relationship between social competence and emotional maturity among adolescents studying in secondary and senior secondary schools of Yamuna Nagar district, Haryana, and to determine whether male and female adolescents differ significantly on either variable. A sample of one hundred adolescents, aged thirteen to eighteen years, was drawn through stratified random sam-pling, balanced across gender, school management, and locale. Data were gathered using two standardised tools — the Social Competence Scale (Mathur & Bhatnagar) and the Emotional Maturity Scale (Singh & Bhargava) — and analysed using Pearson's correlation and the inde-pendent samples t-test at the 0.05 level of significance. Results showed a significant relationship between the two variables, a significant gender difference favouring females on emotional ma-turity, and no significant gender difference on social competence. The paper discusses these findings in light of existing research and offers implications for schools and counsellors.
Doi-[https://doi.org/10.5281/zenodo.21902334]
Academic Achievement of Adolescent Students in Relation to Their Self-Concept
Authors-Sudesh, Dr. Vinod Kumar
Keyword-Academic Achievement, Self-Concept, Adolescents, Secondary School, Locale, Gender.
Abstract-Academic achievement remains one of the central goals of the educational process, and a grow-ing body of research suggests that how a student thinks and feels about themselves — their self-concept — is closely bound up with how well they perform in school. The present paper reports a study undertaken to examine the academic achievement and self-concept of adolescent students studying in secondary schools of Yamuna Nagar district, Haryana, and to determine whether these two variables differ significantly by gender and by locale (rural/urban), and whether they are related to one another. A sample of one hundred and twenty secondary school students was drawn through stratified random sampling from four schools, equally balanced across gender and locale. Self-concept was measured using the standardised Self-Concept Questionnaire de-veloped by P. K. Saraswat, while academic achievement was recorded as students' percentage of marks in the Class 10 board examination. Data were analysed using descriptive statistics, the independent samples t-test, and Pearson's coefficient of correlation. Results showed a highly significant locale-wise difference in academic achievement favouring urban students, no signifi-cant gender difference in either academic achievement or self-concept, and a low but positive correlation between academic achievement and self-concept. The paper discusses these findings against existing literature and offers implications for schools and educators.
Doi-[https://doi.org/10.5281/zenodo.21902621]
Internet Addiction among Senior Secondary School Students: A Comparative Study across Gender, Locality, and Academic Streams
Authors-Dr. Sushma Rani, Ms. Manju Yadav
Keyword-Internet addiction, adolescents, senior secondary students, digital behaviour, gender, locality, academic stream.
Abstract-The Internet has become an indispensable component of contemporary education and communi-cation. Although it facilitates learning, information sharing, and social interaction, excessive Internet use among adolescents has emerged as a growing educational and psychological con-cern. The present study investigated differences in Internet addiction among senior secondary school students with respect to gender, locality, and academic stream. A descriptive survey method was employed to collect data from 90 senior secondary school students using a stand-ardized Internet Addiction Scale. The collected data were analyzed through descriptive statistics (Mean and Standard Deviation) and inferential statistics (independent samples t-test). The find-ings indicated significant differences in Internet addiction between male and female students, rural and urban students, and Arts and Science students. Male, rural, and Arts students exhibited comparatively higher levels of Internet addiction. However, no statistically significant differ-ences were observed between Arts and Commerce students or between Science and Commerce students. The findings suggest that demographic and educational variables influence Internet use patterns among adolescents. The study recommends integrating digital literacy, counselling services, and awareness programmes into school education to promote healthy and responsible Internet use among students.
Doi-[https://doi.org/10.5281/zenodo.21943568]
Turn-Taking and Classroom Participation: Implications for Language Learning in Nigeria
Authors-Adeleye C. B, Akinwamide T. K.
Keyword-Classroom participation, Classroom interaction, Language learning, Nigeria, turn taking.
Abstract-Department of art and Language Education, Ekiti State University, Ado-Ekiti Abstract. This article examines turn-taking and classroom participation and discusses their implications for language learning in Nigeria. It focuses on turn-taking patterns, strategies used by teachers and learners to manage speaking turns, the relationship between turn-taking and classroom participation, and the implications of turn-taking for language learning. The article adopts a qualitative review approach and draws on existing studies of classroom interaction, turn-taking and language learning. The review shows that teacher-to-student, student-to-teacher, teacherto-class and student-to-student patterns shape the distribution of speaking opportunities in the classroom. It also identifies strategies such as asking questions, expressing opinions, using fillers, managing interruptions and sustaining turns as important resources for effective interaction. The article further establishes that equitable and well-managed turn-taking promotes active classroom participation by providing learners with greater opportunities to speak, negotiate meaning and practise the target language. In the Nigerian context, effective turn-taking supports communicative competence, learner confidence and more inclusive language-learning environments. The article concludes that teachers need to deliberately manage speaking turns and employ interactional strategies that increase learner participation and create meaningful opportunities for language use.
Doi-[https://doi.org/10.5281/zenodo.21992600]
Role of MOOCs in Developing Critical Thinking Skills among Secondary School Students: A Systematic Review
Authors-Leju Vidosh, Dr. Reni Francis
Keyword-MOOCs, critical thinking, secondary education, SWAYAM, digital learning, school education, India, systematic review, online learning.
Abstract-Massive Open Online Courses (MOOCs) have emerged as an important component of digital education because they provide learners with flexible access to structured learning resources, multimedia content, discussion activities, formative assessment and opportunities for self-directed learning. In India, the development of SWAYAM and the broader digital-education ecosystem has increased the policy relevance of MOOCs, although much of the empirical literature has focused on higher education rather than secondary schooling. This systematic review examines the potential and documented role of MOOCs in developing critical thinking skills among secondary school students, with particular attention to Indian educational contexts and literature published between 2020 and 2025. The review adopts a PRISMA-informed systematic literature-review approach and synthesises evidence concerning digital learning, MOOCs, critical thinking, learner autonomy, interaction, problem solving and reflective learning. The reviewed Indian literature indicates that MOOCs can support critical thinking when courses incorporate inquiry-oriented activities, problem solving, discussion, reflection, formative assessment and opportunities for learners to evaluate information rather than merely consume instructional content. However, the evidence specific to secondary school students remains limited compared with evidence from higher education. Major barriers include unequal digital access, teacher preparedness, learner motivation, language issues, limited interaction and the risk of passive video-based learning. The review concludes that MOOCs should be regarded not as a replacement for teachers but as a technology-supported pedagogical environment that 1 can strengthen critical thinking when appropriately designed and integrated with classroom instruction. The paper proposes a MOOC-based critical-thinking framework for Indian secondary education and identifies priorities for experimental, longitudinal and school-level research.
Experimental Investigation of Heat Transfer Enhancement in Plate Heat Exchangers Using Hybrid Nanofluids
Authors-Dr. Arun Jacob, Sreeraj S, Razeek A, Renjith T
Keyword-Plate Heat Exchanger, Hybrid Nanofluid, Heat Transfer Enhancement, Pressure Drop, Graphene-TiO₂, MWCNT-ZnO, Thermal Performance
Abstract-Plate heat exchangers are extensively used in thermal systems because of the compactness and high thermal efficiency of such apparatus. In this experiment, the thermal-hydraulic behavior of the corrugated plate heat exchanger is examined based on hybrid nanofluids. Three kinds of hybrid nanofluid are prepared consisting of the following: Graphene-TiO₂/water, MWCNT-ZnO/water-EG, and MWCNT-Al₂O₃/water with volume fractions ranging between 0.05% to 2.0%. These nanofluids are investigated in the range of Re equal to 100-5000. From the results, it can be concluded that hybrid nanofluids improve significantly heat transfer performance when being compared with base fluids. For instance, at the volume fraction of 0.5%, Graphene-TiO₂/water hybrid nanofluids improve up to 21.6% the heat transfer coefficient. Also, the convective heat transfer coefficient increased by 17% in MWCNT-ZnO hybrid nanofluid at the same volume fraction. But, the penalty of pressure drop and increased friction factor is observed with rising nanoparticle concentrations. PEC (performance evaluation criterion) index is utilized to determine the efficiency of hybrid nanofluids.
Doi-[https://doi.org/10.5281/zenodo.22040263]
The Future of Digital Commerce and Consumer Trust
Authors-P.Uma kumari, Dr.A.Parameshwari
Keyword-Digital Commerce, Consumer Trust, Agentic Commerce, Artificial Intelligence, Personalization-Privacy Paradox, Trust Calibration
Abstract-Rapid digital commerce evolution as a result of artificial intelligence use, agency, and fragmented customer journeys necessitates an alteration in the way consumers interact with digital commerce platforms. This paper analyzes the importance of building consumer trust as a core element defining the future of digital commerce. By conducting a qualitative analysis of industry reports, survey data, and research literature, from 2021-2026, the study puts forward the concept of Trust-Calibrated Commerce Framework (TCCF), which consists of the mechanisms of transparency, governance, and personalization-privacy balance. It has been found out that despite 40% of consumers across the world being ready to entrust purchases to artificial agents, the main issue deterring users is trust as 55% believe that AI may steal their identity and make some unauthorized purchases. It has also been revealed that trust calibration mechanisms differ significantly depending on customer segments and young people prefer to deal with AI agents.
Doi-[https://doi.org/10.5281/zenodo.22040390]
Digital Transformation in Human Resource Management and Marketing Strategies
Authors-Maheswari S, Dr Deepa Vijay Abhonkar
Keyword-Digital Transformation, Human Resource Management, Marketing Strategy, Artificial Intelligence, HR Analytics, Strategic Integration
Abstract-Integration of digital transformation, human resource management, and innovative marketing approaches reflects the paradigm shift in today’s organizational practices. The current paper provides the conceptual framework, according to which the integration of digital HRM technology and marketing innovations is able to bring synergic effects within organization. The research methodology is based on the systematic analysis of AI-based HR analytics, automated digital marketing, and the integration of performance measurement systems. The results prove that the use of holistic digital integration strategy brings significant improvements in the effectiveness of an organization in comparison with segmented applications. The research approach is based on the combination of literature review, benchmarking among 23 multinationals and comparative analysis of digital transformation strategies based on the integration/non-integration model. As a result of the research, it was found out that implementation of integrated digital HR and marketing strategies leads to the improvement of employee retention by 17.6%, the increase in the accuracy of revenue forecasting by 22.5% and 110.7% improvement of marketing campaign ROI.
Doi-[https://doi.org/10.5281/zenodo.22040513]
Understanding Pramāṇa in the Indian Knowledge System: Sources, Validation, and Contemporary Relevance
Authors-Soumi Hazra, Dr. Newtan Biswas
Keyword-keywords: Pramāṇa, Indian epistemology, pramā, apramā, perception, inference, testimony, Indian Knowledge System, knowledge validation
Abstract-Abstract. The Indian knowledge tradition has developed a sophisticated epistemological framework for understanding the nature, sources, and validity of knowledge. At the centre of this framework is the concept of pramāṇa, generally understood as a valid means or source of knowledge. Indian philosophers were concerned not merely with what is known, but also with how knowledge is acquired, what makes cognition valid, and how valid knowledge can be distinguished from error. The theory of pramāṇa therefore constitutes an important foundation of Indian epistemology. Different philosophical schools developed distinct classifications of pramāṇas, reflecting their divergent philosophical commitments. The Sāṃkhya and Yoga traditions generally recognize three—pratyakṣa (perception), anumāna (inference), and śabda (verbal testimony); Nyāya recognizes four by additionally accepting upamāna (comparison); while Advaita Vedānta recognizes six, adding arthāpatti (postulation) and anupalabdhi (non-apprehension). Buddhist epistemologists such as Dignāga and Dharmakīrti accept perception and inference as the two reliable means of knowledge, whereas Cārvāka gives primacy exclusively to perception. This paper examines the concept and forms of pramāṇa, its role in the validation of knowledge, its interpretation across philosophical traditions, and its possible application to contemporary knowledge practices. The paper argues that the epistemological concern underlying pramāṇa remains highly relevant to contemporary knowledge systems, particularly in developing critical, evidence-based, and responsible approaches to information.
Doi-[https://doi.org/10.5281/zenodo.22327123]