All of the articles in this proceedings volume have been presented at the International Conference on Intelligent Systems and Digital Transformation, ICISD 2025 during 5th and 6th May 2025 in SRM Institute of Science and Technology, Vadapalani Campus, Chennai. These articles have been peer reviewed by...

Low-flying aircraft pose significant challenges to radar systems, making them impossible to detect. To address these risks, it is critical to detect and identify such aircraft in real time. This work presents an approach for aircraft detection and classification using acoustic signatures. The acoustic...

Reserve Bank of India decides on the interest rate at which it will be lending or borrowing money from other commercial banks. These interest rates are called as Repo rates. There is general consensus that when the repo rate decreases the price of goods and commodities like gold increases, and when repo...

Unmanned Aerial Vehicles (UAVs) have gained global attention across various industries due to technical advancements in system versatility and efficiency. However, achieving accurate and reliable navigation remains a critical challenge, particularly in GPS-denied environments. This paper explores key...

The subtypes of Non-Hodgkin s lymphoma are very important in analyzing the right treatment plans that would be selected and to know the improvement in the outcomes of patients. Nevertheless, addressing the standard histopathological diagnosis is time consuming and subjective. The paper solutions to the...

As technological capabilities of conversational AI models grow so does their susceptibility to cyber threats which in turn raise critical security concerns. The potential of LLMs in aiding cyberattacks is diminutive but never zero. The study extends prior work by evaluating multiple LLMs on their effectiveness...

Science misinformation on platforms like YouTube poses significant challenges to public understanding, necessitating reliable evaluation tools. This project presents TriVerBERT-LLM, a multimodal framework designed to assess the credibility of YouTube video transcripts through logical fallacy detection,...

Cryptocurrency arbitrage offers profit opportunities through price discrepancies across exchanges, but identifying viable paths is challenging due to the volatility, fees, and complexity of multi-token ecosystems. This work proposes a scalable and interpretable GNN-based framework using GraphSAGE with...

Cyberbullying on social media comes with serious challenges that require advanced and interpretable detection mechanisms. We present TriBERT-X, a potent explainable model of transformers, uniting the strengths of ALBERT, RoBERTa, and DistilBERT through a unique concatenation-ensembling strategy. The...

Digital governance aims to enhance the efficiency, transparency, and inclusiveness of public administration through technology. However, current systems often face limitations such as centralized authority, limited civic participation, opaque decision-making, and the absence of intelligent validation...

Traffic accidents are the main cause of deaths, disabilities, and hospital visits in the whole country without stopping. This research is a deep study aiming to choose the best predictive models that can predict traffic accidents. Therefore, cars, whose drivers are essential in managing road safety,...

The healthcare sector is increasingly utilizing drone technology and artificial intelligence (AI) to enhance disease prediction, manage medical supplies, and deliver free healthcare services targeting major causes of death such as heart disease, stroke, and diabetes. These chronic conditions pose serious...

Facial skin tone classification plays a vital role in dermatological AI and the beauty tech industry, enabling personalized skincare, inclusive representation, and health monitoring. Accurate tone detection supports product customization and helps reduce bias in computer vision systems. However,...

Coal miners encounter significant dangers because of the challenging conditions they face underground, necessitating strong safety protocols. This endeavour introduces a Smart Safety Helmet powered by IoT, with Arduino serving as the primary controller, designed for instant monitoring and emergency notifications....

Gait abnormalities can significantly impact an individual’s mobility and independence, necessitating timely assistance and intervention. Gait analysis plays a crucial role in healthcare and assistive technology, helping identify mobility impairments and providing necessary support. However, existing...

Breast cancer remains one of the most prevalent and life-threatening diseases affecting women worldwide. Early diagnosis plays a critical role in enhancing patient survival outcomes. While histopathological analysis—examining tissue samples under a microscope—is the gold standard for diagnosis, it is...

Medicinal plants are an essential part of traditional medicine, but their effective utilization is restricted by limitations in proper identification and authenticity. This paper introduces a mobile app that utilizes high-end image recognition APIs to recognize medicinal plants from photos uploaded by...

In the realm of artificial intelligence (AI), optimizing transformer-based large language models (LLMs) has gained significant traction. These models are pivotal in enhancing AI capabilities for various applications, including health and wellness support. Specifically, they can assist in addressing body...

This research intends to improve the user experience and simplify travel planning by creating an AI-driven Tourism Recommendation System with GPT-3.5 technology, acting as a standalone travel guide. In contrast to conventional NLP-based systems, this solution includes Kernel APIs to provide real-time,...

In today’s fast-changing human resource management (HRM) landscape, organizations are increasingly adopting advanced technologies to improve decision making and workforce efficiency. This study explores how data analytics and generative artificial intelligence (GenAI) are transforming Modern Human Resource...

Most of the crop supply systems in urban areas face multiple challenges such as food waste, limited access for small scale producers and excessive costs. This model covers all agri-producers- both farmers and home growers in urban context. This paper presents Urban AgriFlo an all-in-one AI powered platform...

This paper introduces AIVA (AI-powered Interview Verbal Analysis), a novel system that combines advanced speech-to-text technology with natural language processing (NLP) to automate interview analysis. AIVA integrates a fine-tuned Whisper model optimized for Indian English accents with a large language...

The current fast-moving agricultural environment presents multiple difficulties to farmers because of their restricted access to current precise information. The lack of information creates sustainable damage to agricultural production since crop yields suffer and farmers risk their long-term farming...

The Vision-to-Voice system transforms the comic books stories into engaging audiobooks. The system uses Bootstrapped Language-Image Pretraining (BLIP) for feature extraction and Optical Character Recognition (OCR) for text extraction. Major visual aspects and text in comic panels are detected and translated...

This comprehensive report presents an in-depth analysis of our groundbreaking research on improving Natural Language to SQL (NL2SQL) conversion through the application of modern Large Language Models (LLMs). We meticulously fine-tuned and rigorously evaluated five distinct language models—T5 (serving...

Timely identification of tomato leaf diseases is vital for ensuring agricultural productivity and sustainability. This study presents an innovative ensemble deep learning framework that integrates EfficientNet and Vision Transformer (ViT) to achieve precise and scalable classification of tomato leaf...

IntelliRec is an intelligent FAQ recommendation system designed to enhance placement preparation by delivering company-specific frequently asked questions (FAQs). Traditional question banks are often static and fail to adapt to evolving recruitment patterns, leading to redundant and inefficient study...

Tick-borne Borrelia burgdorferi caused Lyme di1sease remains a major public health concern due to its potential to cause serious complications if left untreated. Traditional diagnosis methods such as ELISA and Western Blot are hindered by late reports and false negative results, particularly in the situation...

In the current fast-paced business landscape, organizations encounter significant challenges related to employee retention and sustainable development. This paper introduces the Smart HR Sustainability System, an AI-powered analytics framework designed to forecast employee turnover, improve retention...

This research introduces a novel framework titled “Analysis of Financial Data to Plan and Forecast for Resource Allocation Using Double Machine Learning”, which aims to overcome the limitations of traditional forecasting methods in financial analysis. Existing models such as Arima and Long Short Term...

Precision irrigation plays a crucial role in optimizing water usage and improving crop yield in modern agriculture. This paper presents an IoT-based precision irrigation system designed specifically for eggplant and tomato cultivation. The system integrates soil moisture sensors, temperature and humidity...

For the sake of protecting lives and property and an efficient traffic management, in the domain of current transportation accurate prediction of the traffic signs is extremely important. We present a novel Traffic Sign Detection and Navigation System for the Visually Impaired using the Artificial Intelligence...

EHR systems require the integration of Blockchain with Homomorphic Searchable Encryption technology to achieve secure protection and privacy management of data access. Blockchain technology stands out as an appropriate platform for healthcare data safety because it maintains both decentralized control...

The presented project uses the XGBoost algorithm along with a Hybrid Long Short-Term Memory network to develop an intricate system of land mine detection. The system aims to enhance active mine identification accuracy via simulated data analysis. Hybrid uses both strong classification performance of...

The rising necessity for intelligent and sustainable power management in modern electricity networks has led to the evolution of smart grids, supported by sophisticated data analytics and communication frameworks. This research introduces a deep learning-based method designed to anticipate energy requirements...

Effective quality control in luxury leather goods manufacturing requires objective assessment methodologies. This paper introduces an automated leather bag quality analysis system with a three-layer architecture: input for image preprocessing, processing for defect detection, and output for results visualization....

Religious places attract huge crowds and the density of massive crowds which are measured by manual crowd counting and monitoring by surveillance are not effective. Light variations in religious place premises present a great challenge to modern machine learning (ML) techniques as well as traditional-based...

Online comment sections, especially on social media platforms like Reddit, offer rich opportunities for public discussion—but they also suffer from an overwhelming volume of low-quality, repetitive, or toxic content. While previous research has largely focused on sentiment analysis or the detection of...

Blood banks play a crucial role in ensuring safe and timely blood transfusions, making them essential to healthcare systems. However, traditional blood bank management often relies on manual processes, leading to inefficiencies, delays, and logistical challenges, especially during emergencies. This paper...

Heart disease is a leading cause of mortality worldwide, highlighting the urgent need for early detection and timely medical intervention. This study introduces an integrated system that leverages machine learning, specifically a Random Forest classifier, to predict heart attack risk based on structured...

Hiring is often very hard for small businesses because they don’t have enough people, their budgets are tight, and they get a lot of applications. Traditional Applicant Tracking Systems (ATS) work well, but they are often too expensive and complicated for small businesses to use. An AI-powered ATS made...

One well-known computer vision algorithm that makes applications like surveillance and driverless cars possible is object detection. This study introduces an OpenCV-based real-time item tracking and detection system based on the MobileNet SSD model. It has a tracking module for ongoing tracking and makes...

In the current digital age, ensuring the security and integrity of data has never been more vital. Traditional cloud storage services are easy to use and, while great in many use cases, these services can expose information, restrict access to the data, and are susceptible to a single point-of-failure...

Efficient inventory management is critical in e-commerce, where inaccurate demand forecasts can lead to costly stock outs or excess inventory. This study presents a scalable, machine learning-based framework to optimize warehouse operations. Historical sales data—including product/store identifiers,...

This study uses a One-Dimensional Convolutional Neural Network (1D-CNN) and deep learning to identify obstructive sleep apnea (OSA) from electrocardiogram (ECG) signals. The PhysioNet Apnea-ECG database's ECG recordings were divided into 60-s segments and categorized as either apnea or non-apnea....

The continuous advancement of cyber threats has diminished the effectiveness of conventional network intrusion detection methods. This research proposes an innovative intrusion detection strategy that combines frequency-domain analysis with machine learning to detect harmful network traffic. By applying...

Analyzing employee attrition helps us understand core issues employees may be facing, like challenging workloads, inadequate compensation and psychological distress. Our project tries to understand reasons for attrition in healthcare—an industry where it is critical that the employees function efficiently,...

Early identification of wildfire outbreaks is crucial to minimize ecological damage and reduce the risk to human life and property. Conventional satellite-based detection techniques often lack the temporal resolution needed for rapid response. This study presents a cost-effective and deployable wildfire...

Government fund management is a very important function that has to be efficient, transparent, and accountable. However, with the conventional systems, inefficiency, untransparency, and high cost that go hand in hand with delays and possible abuse are intrinsic features. Blockchain presents a hopeful...

Tracking student attendance is often a time-consuming task, particularly in large classroom settings, where monitoring each individual's presence becomes challenging. In some instances, students may even impersonate their peers to mark attendance in their absence. To address this issue, the proposed...

Traditionally, predictions for cricket matches have concentrated on team performance metrics, frequently neglecting the effects of individual player interactions. This study introduces a new player-vs-player method for predicting T20 innings scores, utilizing historical ball-by-ball data from IPL games....

The presence of diseases in plant leaves poses a serious challenge to agricultural productivity, often leading to considerable financial losses. Identifying and diagnosing these diseases at an early stage is essential for effective crop management. This study introduces an advanced method for automated...

The increasing demand for adaptive learning systems has highlighted the limitations of traditional, one-size-fits-all educational approaches. StudyPalz, a personalized learning platform has been crafted to help students study more effectively by pinpointing their weak spots and providing tailored resources...

In contemporary businesses, managing the increasing number of issues and complaints by hand frequently leads to delays and inefficiencies. To solve this, we suggest an AI-powered system that automatically analyzes and ranks text-based inputs from emails, tickets, and feedback using Natural Language Processing...

With the increasing volume of structured data across enterprises, efficient data retrieval and analysis have become critical challenges. This paper presents an AI-powered document chatbot that enables seamless data access using a Retrieval-Augmented Generation (RAG) architecture. The chatbot leverages...

The integration of artificial intelligence and machine learning into trading is significantly transforming financial markets, often surpassing traditional methods in terms of efficiency and accuracy. This paper presents a comprehensive study that encompasses a systematic literature review of algorithmic...

In today’s era of advanced technology, changes in our daily habits have significantly contributed to a rise in health-related issues, particularly heart diseases. With millions of deaths reported annually due to cardiovascular complications, it has become crucial to focus on early diagnosis and public...

With the increasing shift to digital financial systems, detecting credit card fraud has become an ongoing challenge for banks and institutions. This paper introduces a comprehensive machine learning-based fraud detection system that combines oversampling through SMOTE with ensemble classification models....

Road traffic accidents are a critical public safety issue globally, causing substantial loss of life, serious injuries, and economic consequences. Accurate prediction of accident severity can enhance preventive measures, emergency response, and policy formulation. This study proposes a machine learning...

Accurate and precise classification of cardiac arrhythmias is essential for early diagnosis and prevention of life-threatening cardiac conditions. This research introduces an innovative two-stage deep learning approach for classifying arrhythmias, utilizing Dual-Tree Complex Wavelet Transform (DT-CWT)...

In the era of rapid digital communication, social media platforms like Twitter (now X) serve as critical channels for public opinion and discourse. This project presents a comprehensive sentiment analysis pipeline that integrates both lexicon-based and transformer-based natural language processing (NLP)...

Businesses must practice effective inventory management to satisfy client requests and keep prices down. Preventing instances where an oversupply or stock out largely depends on accurate demand forecasting. Inventory demand forecasting systems have advanced considerably by integrating machine learning...

Diabetic Retinopathy (DR) is the most common cause of vision loss globally, prompt detection is essential for efficient treatment and intervention. High-resolution context of the retinal fundus images is problematic from viewpoints of resolution constraints, loss of important lesion information, and...

Deep learning-based image classification models have been established as a potent approach in the medical imaging domain as they are able to provide higher-level accuracy for disease diagnostic tasks. Convolutional Neural Network (CNN) architectures have shown great promise from among these in parsing...

Skin lesion detection using machine learning has emerged as a crucial tool in early diagnosis and treatment planning for skin cancer, particularly melanoma. This research focuses on implementing multiple machine learning algorithms, including Random Forest, Support Vector Machine (SVM), and Decision...

Influencers often transcend single content niches, impacting diverse audiences across categories. This study investigates cross-category influencers to understand their role in engagement dynamics, brand collaborations, and network influence. Unlike traditional approaches focused on follower counts,...

Food wastage and hunger are two major problems globally, most famously seen in most developing countries like India, where they are the norm. The paradox of excess food wastage and extensive hunger is a major impetus for creative solutions. Zero Hunger, a trailblazing web-based food donation platform,...

As everything is going digital nowadays, online activity is a treasure trove of information to analyze and learn what humans love and how they behave. In this study, deep learning techniques are used to derive behavioral insights based on study questions to gauge relationship compatibility. Utilizing...

Technology upgrades our lifestyle, working culture, and day-to-day activities in this era, Artificial intelligence is playing a vital role in making things easier, embraces human-computer interactions and enhances human knowledge to unprecedented levels. The aim is to develop a small-scale voice-based...

The implementation of large language models (LLMs) on edge devices encounters three major challenges: the requirement of high computational resources, threats to users data privacy, and degradation in performance resulting from model compression in environments with limited resources is a significant...

Counterfeit money is a serious threat to economic stability and public confidence. This paper introduces an integrated Fake Currency Detection System based on Convolutional Neural Networks (CNN) and Optical Character Recognition (OCR) for detecting fake Indian banknotes. In contrast to conventional methods...

Hand gesture recognition offers a natural and intuitive mode of human computer interaction by enabling users to control digital systems through physical movements. However, many existing approaches are constrained by the need for complex hardware, large annotated datasets, or computationally intensive...

Facial emotion detection has emerged as a critical area in human-computer interaction, healthcare, and security applications. Traditional methods often relied on handcrafted features and single-label classification, which lacked depth and flexibility. This paper proposes a deep learning-based model utilizing...

Accurate crop yield prediction is vital for enhancing food security and supporting data-driven agricultural planning. Existing models often fail to capture the intricate relationships between environmental conditions, farming practices, and crop responses. This study presents a novel deep learning-based...

In digital environments, social networks are profoundly shaped by the actions of influential nodes, such as celebrities, content creators, and highly engaged users. Their interactions—liking, sharing, commenting, visiting pages, and purchasing products—serve as catalysts for engagement cascades that...

This study aims to enhance the performance and equity of Peer-to-Peer (P2P) lending systems by applying machine learning techniques for borrower assessment and interest rate forecasting. The proposed framework operates in two phases. Initially, a Random Forest Classifier is used to assess borrowers’...

Smart farming requires smart systems capable of making timely and accurate suggestions for crop choice, fertilizer usage, and detection of plant disease. This article presents AgroAware, a web-based system that utilizes machine learning and deep learning techniques to facilitate precision agriculture....

Heart disease continues to be the primary cause of death globally. It highlights the importance of reliable and efficient diagnostic methods to reduce its impact. This paper presents a state-of-the-art ensemble-based machine learning architecture for predicting heart disease, aiming to enhance diagnostic...

The proliferation of deepfake media stirred grave doubts over media authenticity threatening digital trust and social coherence. This paper proposes a new technique of deepfake detection from images and videos through the use of a hybrid of MobileNet spatial feature extraction and Long Short Term Memory...

In the landscape of modern digital commerce, user- generated reviews are pivotal in shaping consumer preferences, influencing purchasing behavior, and contributing to brand perception. However, the infiltration of fraudulent or misleading reviews undermines the credibility of such platforms. This paper...

It utilizes machine learning prognostics in predicting genetic diseases by taking large sets of computational algorithms that process huge genetic data. These models attempt to forecast the probability of acquiring certain types of genetic diseases with the help of gene markers, ancestral history, environment,...

Prison security needs fast and accurate systems to detect weapons and identify people in real time. Depending only on human guards is not always reliable, as they can miss threats or react slowly. In this paper, a smart surveillance system that uses deep learning to detect weapons and recognize prisoner...

Personalized grocery recommendation systems enhance customer experience and retail performance. While traditional methods like collaborative filtering struggle with dynamic preferences, modern deep learning and hybrid systems achieve superior accuracy and personalization. However, Traditional recommendation...

Image restoration and enhancement have significantly progressed in restoring degraded images with haze, missing areas, and structural ill-clarity. Conventional restoration techniques are not able to retain global structures and fine textures simultaneously, resulting in artifacts and loss of quality....