Volume 14 | Issue 4

 Abstract

From the Kothari Commission to NEP 2020: Continuity and Change

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

From the Kothari Commission to NEP 2020: Continuity and Change

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

A case report details the physiotherapy rehabilitation procedure of a 48 yrs old man having multiple traumatic fractures from a fall injury, fracture is seen involving L3 and L4 vertebral bodies with about 10-20% reduction in vertebral body height. Fracture seen involving the posterior vertebral elements at these levels. The note is made of a fracture involving the right proximal femur. linear undisplaced fracture lines, involving right sided inferior pubic ramus. Fracture is seen involving D7 to D11 right sided transverse process. Fracture is seen involving L2 to L4 right sided transverse process. Left sided 8th rib fracture is seen in its posterior aspect. Fracture involving proximal humerus is seen. After open reduction and internal fixation surgery, an 8-12-week physical treatment regimen was put into place. Exercises for passive and active range of motion, isometric and progressive resistance training, and gait training were provided. The rehabilitation goals were pain relief, increased range of motion, muscle strength, flexibility, endurance, and functional independence. Pain levels, range of motion, muscle strength, and general function all significantly improved between pre- and post-rehabilitation evaluations. Early mobilization and structured physical therapy were crucial in achieving these outcomes, highlighting the importance of tailored rehabilitation protocols for post operative recovery.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Polytrauma,physiotherapy rehabilitation,lisfranc fracture,pelvic fracture, functional independence measure, barthel index.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents Dr.AI, a comprehensive AI-driven multimodal healthcare chatbot platform designed to democratize medical guidance through advanced natural language processing, Retrieval-Augmented Generation (RAG), and real-time emergency response capabilities. Dr.AI integrates Meta's llama-4-scout-17b-16e-instruct language model accessed via the Groq LPU inference engine, a curated ChromaDB vector store of 3,760 medical knowledge chunks, OpenAI Whisper for speech-to-text transcription, and Google Text-to-Speech (gTTS) for audio output. The system supports over 16 languages including Hindi, Hinglish, Marathi, Tamil, and Telugu, enabling voice-based and text-based medical consultation across diverse linguistic populations. Key features include a 5-step AI symptom checker, emergency SOS with real-time GPS hospital finder, medicine reminder scheduling, family health profile management, AI-powered medical report analysis for X-rays and lab results, webcam image capture, and a downloadable PDF health report. The platform is deployed on Render (backend) and Vercel (frontend) with Firebase Firestore for persistent data storage. Experimental results demonstrate that RAG-enhanced responses achieve approximately 85-90% factual grounding accuracy compared to 65-70% without RAG on domain-specific medical queries. The paper describes the system architecture, RAG pipeline design, deployment infrastructure, multilingual capabilities, and emergency detection mechanisms, establishing Dr.AI as a scalable, accessible, and clinically aware AI health companion.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

speech recognition, Whisper AI, Llama 4 Scout, medical image analysis, text-to-speech, gTTS, disease detection

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Artificial Intelligence has significantly transformed modern software engineering practices, particularly in the domain of automated software testing. Large Language Models (LLMs) are increasingly used to generate automated test cases by understanding source code, system requirements, and documentation. Although these models can generate test cases quickly and improve testing productivity, they often suffer from issues such as hallucination and incorrect assertions. Hallucination refers to the generation of logically incorrect or non-existent information that appears syntactically valid but does not accurately represent the expected behavior of the software system. In the context of test case generation, hallucinated outputs may include incorrect expected results, fabricated APIs, irrelevant assertions, or invalid boundary conditions. These problems reduce the reliability of AI-generated test cases and require manual verification by developers.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Automated Test Case Generation, Large Language Models (LLMs), Software Testing, Test Automation, Hallucination in AI, Assertion Validation, Mutation Testing.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In recent years, the healthcare sector has generated large volumes of patient data that can be effectively utilized to predict diseases at an early stage. This project focuses on developing a Disease Risk Prediction System using Machine Learning techniques to analyze patient health data such as age, lifestyle habits, medical history, and clinical parameters. The proposed system predicts the risk level of diseases such as diabetes, heart disease, or hypertension and provides early warnings to patients. This system helps in preventive healthcare, reduces late diagnosis, and supports doctors in clinical decision-making. The model improves healthcare efficiency by enabling timely medical intervention and personalized care.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning, Disease Risk Prediction, Healthcare Analytics, Early Diagnosis, Predictive Modeling, Diabetes Prediction, Heart Disease Prediction, Hypertension Detection, Patient Data Analysis, Clinical Decision Support System, Preventive Healthcare, Data Mining, Artificial Intelligence in Healthcare, Health Monitoring, Risk Assessment

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The aim of the research was to determine how gamified learning platforms impact the academic performance of college students (i.e. performance) and their participation in a college environment compared to traditional online environments. An approach of mixed methods was taken in which data was collected within a time span of 12 weeks on 50 undergraduate students pursuing Management and Computer Science courses. The group of students who took part in this study was divided into the Experimental group or Control Group. In the case of the Experimental Group, the gamified platforms employed at this time were Kahoot! Quizizz, Duolingo, and Classcraft. In the Control Group, non-gamified platforms that were used at this time were Google Classroom and Moodle. The quantitative findings have shown that Experimental Group scored significantly higher on an average post-test result as compared to Control Group (Experimental: 18.4 pts; Control: 10.3 pts). The Experimental Group had much greater indices ofstudent engagement in the areas ofstudent attendance, activity completion and activity participation as compared to Control Group. The Experimental Group did not lose their drive throughout the 12 weeks and they only experienced a slight decline in engagement. In the Control Group, the participation reduced significantly during the study. This study could be useful in finding the usefulness of gamified methodology within the framework of traditional, full-time degree program courses in an education environment that is quickly transforming due to digitization.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

e-learning, gamified, non-gamified, gamification, education, kahoot!, Quizizz, Duolingo, Classcraft, platforms, teachers, technologies.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Agriculture remains a primary livelihood for a large portion of the global population; however, farmers continue to face challenges such as price uncertainty, market fluctuations, and limited trust in existing trading systems. In many cases, traditional supply chains depend on intermediaries and centralized databases, which can reduce farmers' profit margins and restrict transparency. Issues such as data inconsistency and lack of reliable market information further complicate decision making. This paper presents Krishi-Unnati, a digital platform designed to improve transparency and support better decision making in agricultural trading. The system integrates blockchain technology, artificial intelligence, and mobile computing within a hybrid architecture. Transactional data, including trade records, is stored on a blockchain ledger to ensure integrity, while other operational data is maintained off-chain to improve efficiency. The platform provides two main functionalities. First, a crop price prediction module based on Random Forest Regression is used to estimate future market prices. Second, a buyer recommendation engine analyzes historical transaction data to suggest suitable trading partners. Together, these components assist farmers in making more informed choices regarding crop sales and market participation. The mobile application is developed using React Native, with a backend built on Node.js, Express.js, and MongoDB. System evaluation indicates improvements in transparency, access to market information, and overall decision support. The proposed approach contributes toward a more reliable and data-driven agricultural trading ecosystem.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blockchain, Smart Agriculture, Price Prediction, Buyer Recommendation, Supply Chain Transparency, Smart Contracts, Agriculture Technology

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing integration of Artificial Intelligence (AI) into cyber security systems has transformed the way digital threats are detected, analyzed, and mitigated. While AI enhances efficiency and accuracy in cyber defense mechanisms, it simultaneously introduces complex legal and ethical challenges. This paper critically examines the implications of AI-driven cyber security, focusing on issues such as data privacy, algorithmic bias, accountability, and lack of transparency. AI systems rely heavily on large datasets, raising concerns regarding unauthorized data collection and potential misuse. Moreover, the opaque nature of AI algorithms often complicates legal accountability when automated decisions result in harm. Ethical concerns such as excessive surveillance, discrimination, and erosion of individual rights further intensify the debate. Existing legal frameworks remain inadequate to fully regulate AI technologies, highlighting the urgent need for comprehensive and adaptive policies. This study emphasizes the importance of balancing technological advancement with ethical responsibility and legal compliance. It concludes that a multidisciplinary approach involving legal experts, technologists, and policymakers is essential to ensure that AI-driven cyber security operates within a framework that respects human rights and promotes transparency, fairness, and accountability.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Cyber Security, Legal Challenges, Ethical Issues, Data Protection, Algorithmic Bias

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

ekuokf/kdkj] Ekkuokf/kdkj guu] lkekftd vlekurk] dkuwu izorZu] cky Je] efgyk fgalk] xjhch] HkkjrA

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Rapid urbanization in India has overwhelmed traditional, reactive complaint systems, leaving ward-level issues like garbage overflow and infrastructure decay unresolved. This paper proposes WardPulse, an AI-driven proactive monitoring system integrating IoT sensors, Computer Vision, and crowdsourced reporting. By utilizing Convolutional Neural Networks (CNN) with MobileNetV2 for image-based defect detection and a geo-tagged dashboard for municipal authorities, the system automates issue classification and prioritization. Experimental results show a 40% reduction in resolution time and significant improvements in citizen engagement. This scalable framework is designed for seamless deployment in Smart City initiatives across both municipal corporations and gram panchayats.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Cities, Computer Vision, MobileNetV2, IoT, Civic Technology, and Urban Infrastructure.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Tomato (Lycopersicum) is one of the most economically important horticultural crops in India, especially in Andhra Pradesh, which accounts for about 18% of the country's tomato production. Tomato plants are greatly affected by many different leaf diseases. These include fungal problems like Early Blight and Septoria Leaf Spot, bacterial issues such as Bacterial Spot, and viral infections like Yellow Leaf Curl Virus and Mosaic Virus. If these diseases are not found and dealt with quickly, they can cause a loss of 40 to 60% of the crop. Traditional ways of identifying diseases by looking at plants, done by agronomists, take a lot of time, are easy to make mistakes, and don't work well when you need to check a large number of crops, especially in rural areas where it's hard to get expert help. This project introduces an automated tomato leaf disease identification system utilizing the Inception V3 deep convolutional neural network architecture combined with transfer learning. The model was trained using the PlantVillage dataset, which includes 10 different categories of tomato leaves and around 18,160 images, all obtained from Kaggle. A two-step transfer learning approach is used: first, features are extracted using the frozen weights from the pre-trained ImageNet model, and then the top 30 base layers are fine-tuned with a lower learning rate. Data augmentation techniques such as horizontal and vertical flipping, random rotation, zoom, and brightness adjustment are utilized to tackle class imbalance and enhance model generalization. The system gets an average accuracy of 97.2% on the test set it hasn't seen before. It also has a precision of 0.923, recall of 0.916, and an F1-score of 0.919 when looking at all classes equally. The trained Keras model is converted into ONNX format to make CPU inference run 30 to 40% faster using ONNX Runtime. An HSV-based leaf detection module filters out non-leaf inputs before making predictions, and it uses a confidence threshold to clearly mark any uncertain results

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Tomato Leaf Disease Detection, Inception V3, Transfer Learning, Convolutional Neural Network (CNN), PlantVillage Dataset, Deep Learning, Data Augmentation, Precision Agriculture

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Cardiovascular disease (CVD) is one of the major causes of mortality around the globe, necessitating the need for early detection that will enable better health care results. Existing diagnostic methods utilize significant clinical expertise along with extensive testing; hence, it becomes time-consuming in detecting diseases. This study proposes the "Deep-Ensemble Cardiovascular Disease Detection System" which uses the power of deep learning to detect CVDs at an early stage. The proposed system incorporates a number of deep learning algorithms to develop an ensemble for better performance in predicting the results. In the data preprocessing process, the dataset is cleaned and balanced to improve model learning. To ensure effective prediction and detection of heart diseases, the ensemble uses the CNN and GRU architectures to capture the patterns from health data. Primarily, rather than classifying patients in two categories, including No Disease and Disease, the categorical prediction is extended to four categories, including No Disease, Low Risk, Medium Risk and High Risk. Also, this categorical prediction is not limited to two categories, namely No disease and Disease but extended to four categories, namely No Disease, Low Risk, Medium Risk and High Risk. First, instead of two categories of predictions, such as No Disease and Disease, the categorical prediction is expanded to four categories, including No Disease, Low Risk, Medium Risk, and High Risk. Second, it does not only involve the prediction of two categories, i.e., No disease and Disease, but the expansion to four categories, including No Disease, Low Risk, Medium Risk, and High Risk. Third, an online cardiovascular disease prediction and an internet-based application are realized. Lastly, a categorization of risk prediction is performed by the system. From experimental results, it can be observed that the proposed deep-ensemble model assures higher levels of accuracy, reliability, and usability compared to conventional machine learning methodologies. The proposed system can be used by healthcare professionals for the purpose of early diagnosis, risk assessment, and preventive treatment planning, which helps improve patient care and reduce mortality.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cardiovascular Disease Detection, Deep Ensemble Learning, CNN-GRU Model, Multi-Class Classification, Risk Prediction, Web-Based Prediction System, Artificial Intelligence, Healthcare Analytics.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The basic thesis of growth is the economic growth of which is required for political, social and economic stability the quality environment normally assumes lower priority in planning proposals and long-term planning. Unlimited exploitation of nature by man disturbed the ecological balance between living and non-living components of the biosphere. The adverse conditions created by man himself threatened the survival not only of man himself but also other living organisms. Due to progress, industries, technology, chemicals, atomic energy, there are a number of industrial effluents and emissions of poisonous gases in the atmosphere and also added solid waste which has lowered the quality of environment. The pollution is a necessary evil for all development. Due to lack of development of culture of pollution control, there has resulted a heavy backlog of gaseous, liquid and solid pollution in our country. Thus, pollution control in our country is a recent environmental concern. There is a race in developed countries to exploit every bit of natural resources to convert them into goods for their use and comfort and to export them to other needy countries. The industrialized countries dump lot of materials in their environment which becomes polluted. The environmental pollution has lowered its quality.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Environmental Degradation, Pollution, Institutional Failure, Conflicts, Energy

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing prevalence of skin diseases worldwide highlights the need for early and accurate diagnosis to prevent complications and ensure timely treatment. Traditional diagnostic approaches rely heavily on dermatologists' expertise and manual examination, which can be time-consuming, subjective, and limited in accessibility. To address these challenges, this project proposes an AI-assisted skin disease diagnosis system integrated with a virtual medical expert and appointment booking facility, leveraging deep learning and advanced image processing techniques to provide efficient, accurate, and user-friendly healthcare support. The system enables users to upload skin images, which are preprocessed using techniques such as denoising, resizing, segmentation, and normalization to enhance image quality. Deep learning models, including Convolutional Neural Networks (CNN), ResNet, and EfficientNet, are utilized to classify a wide range of skin diseases with high accuracy. The model is trained and validated on diverse datasets to ensure robustness and reliability in real-world scenarios. Upon diagnosis, the system provides detailed results along with precautionary measures and basic medical recommendations. To enhance user interaction and accessibility, a virtual medical expert chatbot is integrated into the system. This chatbot offers explanations of predicted conditions, suggests preventive steps, and answers user queries, thereby bridging the gap between patients and medical professionals. Furthermore, the system incorporates an appointment booking feature that allows users to schedule consultations with dermatologists, ensuring a seamless transition from diagnosis to treatment. In conclusion, the proposed system presents a comprehensive solution by combining AI-based diagnosis, virtual consultation, and appointment scheduling into a single platform. Future enhancements may include expanding datasets for better generalization, improving model interpretability through explainable AI, and ensuring data privacy and compliance with healthcare standards. This project demonstrates the potential of integrating artificial intelligence with healthcare services to deliver more accessible, efficient, and patient-centered solution.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

CNN, ResNet, EfficientNet

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Financial awareness is essential in influencing the investment habits of retail investors. Based on theories like the Financial Literacy Theory and the Theory of Planned Behavior, this research describes how awareness, mindset, and perceived control affect investment choices. Through government-supported initiatives like the RBI Retail Direct Scheme, individuals can engage directly in government securities markets without needing intermediaries. This research explores how financial awareness influences retail investors in Bengaluru to adopt government direct retail investment schemes. The study relies on primary data gathered from 120 participants through a structured questionnaire. Statistical methods like percentage analysis, correlation, and regression were used to examine the data. The results show a strong positive correlation between financial awareness and the uptake of these schemes, reinforcing the theoretical notion that increased knowledge results in improved financial decision-making. Nonetheless, elements like insufficient awareness, intricate procedures, and restricted digital literacy persist as obstacles. The research highlights the importance of focused financial education programs, streamlined investment procedures, and digital literacy initiatives to enhance participation in government-supported investment platforms.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Financial Awareness; Financial Literacy; Retail Investors; Investment Behaviour; Government Securities; RBI Retail Direct Scheme; Financial Inclusion; Digital Financial Literacy; Investment Decision-Making; Investor Awareness

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

ABSTRACT:- India is a Socialist Democratic country and its economy is based on agriculture. Land is one of the most important sources of income. Human beings has right to live this land but the ownership of the land varies. Since the early stage did not have any law for the ownership of the land but the time passed then scenario was changed. The legal perspective of land acquisition in India has undergone a remarkable transformation with the exchange of the colonial Land-Acquisition Act of 1894 by the Right to Fair Compensation and Transparency in LARR, 2013.This paper tried to establish the development of land acquisition laws in India and the transformation of new LARR Act. Through an analysis of statutory provisions, judicial interpretation, the study highlights both progressive feature of the act and the persistent gaps in its execution.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

KEY WORDS: -Land, Acquisition, Transformation, Compensation, Legal.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The transformation of traditional libraries into AI-powered smart libraries marks a new era of information accessibility, user engagement, and community development. Artificial Intelligence enables libraries to go beyond being repositories of books; they now serve as interactive, data-driven knowledge hubs that anticipate user needs and provide seamless learning. This transformation benefits both individual users and the wider community, fostering lifelong learning, inclusivity, and innovation.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Library Transformation, Smart Libraries, User Engagement, Community Impact, Library Automation, Information Retrieval, Digital Library Services.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Logistics management, warehouse effectiveness, supply chain, Royal Enfield, inventory control, automobile industry, transportation efficiency, ANOVA, Chi-Square test.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Deep Learning-Based Audio Language Model (ALM) is an intelligent multi-modal audio analysis system designed to enhance situational awareness in complex and high-risk environments. Traditional audio monitoring systems treat Acoustic Event Detection (AED) and Automatic Speech Recognition (ASR) as independent tasks, resulting in fragmented intelligence outputs. This project proposes a unified deep learning framework that jointly analyzes speech and non-speech battlefield sounds to generate structured and actionable intelligence reports. The system classifies acoustic inputs into seven operational categories: Communication, Gunshot, Footsteps, Shelling, Vehicle, Helicopter, and Fighter. For non-speech event detection, YAMNet-- a deep convolutional neural network pre-trained on large-scale audio datasets--is used as a feature extractor to generate high-dimensional embeddings. These embeddings are processed by a custom Keras-based dense classifier trained on a curated Military Audio Dataset. In parallel, Faster-Whisper, an optimized transformer-based speech recognition model with 8-bit quantization, performs multilingual transcription of tactical communications. The core innovation of the proposed model lies in its Fused Intelligence Output mechanism, which integrates event classification results and speech transcripts into a structured Situation Report (SITREP). This fusion reduces cognitive load on analysts by presenting prioritized event summaries, confidence metrics, and contextual insights within a unified dashboard environment developed using Streamlit with Role-Based Access Control (RBAC). By combining deep learning-based acoustic event detection and real-time speech transcription into a single cohesive intelligence framework, the proposed ALM system significantly improves response readiness, accelerates decision-making processes, and demonstrates the potential of multi-modal audio intelligence systems in tactical and security-focused applications.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Edge Computing, Signal Filtering, Integrated Speech Processing ,AI/ML

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The present study is aimed to enhance the knowledge regarding cervical cancer among adolescent girls. On Inclusion criteria, 200 adolescent girls were selected from Khalsa College Girls Senior Secondary School Amritsar, Punjab. Total enumerative sampling technique was used to select the sample. Personal information was assessed by using socio-demographic profile and structured questionnaire was used to assess the knowledge of adolescent girls. Study finding revealed that in the pre-test maximum (97%) had average knowledge, however in post-test (75.5%) had very good knowledge regarding cervical cancer. After comparing between pre-test and post-test knowledge of adolescent girls, the post-test mean score (20.41) was higher than pre-test mean score (13.06) at p<0.01 this showed that structured teaching program had significant effect in increasing the knowledge regarding cervical cancer among adolescent girls. The association of post-test knowledge with age, stream of class, source of information, family history of cervical cancer found statistically significant at p<0.05. Hence it can be concluded that cervical cancer is an important factor to be discussed with the adolescent girls of Khalsa College Girls Senior Secondary School, Amritsar. So, it is important to organize the teaching programme to enhance the knowledge regarding cervical cancer.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Knowledge, cervical cancer, adolescent girls

  License

Creative Commons Attribution 4.0 and The Open Definition