SLIIT Conference and Symposium Proceedings

Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/295

All SLIIT faculties annually conduct international conferences and symposiums. Publications from these events are included in this collection.

Browse

Filters

20257

Advanced Search

Filter by

Settings

Start date: 2025Subject: search.filters.subject.Machine Learning

Search Results

Now showing 1 - 7 of 7
  • Thumbnail Image
    PublicationOpen Access
    A Data-Driven Approach to Predicting Ischemic Heart Disease Risk in Monaragala: Integrating Lifestyle and Symptom Factors with Machine Learning
    (Faculty of Engineering, 2025-09-09) Meddepola, M.A.R.L.; Wickramasinghe, B.M.G.S.T.S.K.
    Ischemic Heart Disease (IHD) remains a leading cause of mortality worldwide and presents a critical challenge in underserved rural areas such as Monaragala, Sri Lanka. Traditional IHD prediction methods predominantly depend on clinical diagnostics like ECGs and blood tests, which are often unavailable or inaccessible in such regions. This study aims to bridge this gap by developing a machine learning-based prediction model that utilizes only lifestyle and symptom-related data, eliminating the need for invasive clinical procedures. A dataset comprising lifestyle habits (e.g., diet, smoking, alcohol use, exercise) and symptom indicators (e.g., chest pain, fatigue, dizziness) was collected via surveys. Feature selection using Logistic Regression identified the top eight most relevant predictors. Five machine learning algorithms, Logistic Regression, K-Nearest Neighbors, Support Vector Machine, Decision Tree, and Random Forest, were trained and evaluated. Among them, the Random Forest model achieved the highest performance with an accuracy of 83.5%, precision of 0.86, recall of 0.78, and F1- score of 0.81, demonstrating strong predictive capability based solely on non-clinical features. In addition, a web-based self-assessment tool was developed to make the model accessible to the public, particularly targeting individuals in rural areas with limited healthcare access. The tool enables users to input basic lifestyle and symptom information and receive a real-time risk assessment. The findings confirm that the model leveraging lifestyle and symptom data can effectively identify individuals at risk of IHD. This approach supports the development of scalable, low-cost, and user-friendly screening tools that can enhance early detection and preventive care, especially in rural and resource-constrained settings.
  • Thumbnail Image
    PublicationOpen Access
    Predictive Modeling for Personalized Cancer Therapy Using Reinforcement Learning
    (Faculty of Engineering, 2025-09-09) Edirisinghe M.M; Gunarathne, J H M S M
    Adaptive therapy is transforming cancer treatment by enabling dynamic, patient-specific interventions that adapt to tumor progression and individual variability. Unlike traditional fixed-dose regimens, adaptive therapy leverages the evolutionary dynamics of tumors to extend treatment effectiveness and delay resistance. Reinforcement Learning (RL), an area of artificial intelligence focused on sequential decision-making, offers a robust framework for optimizing these adaptive strategies. RL can learn optimal treatment policies by interacting with computational models of tumor growth and drug response, continuously adjusting regimens based on observed tumor states, resistant cell populations, and biomarkers. This approach allows for the creation of personalized therapies that maintain long-term tumor control while minimizing toxicity and the emergence of resistance. The integration of RL into predictive modeling for cancer therapy represents a paradigm shift, enabling smarter, safer, and more effective treatments that are dynamically tailored to each patient’s evolving disease. This paper reviews the foundational concepts of adaptive therapy and RL discusses tumor modeling approaches, examines RL algorithms, and addresses current challenges and future directions in the field.
  • Thumbnail Image
    PublicationOpen Access
    Enhancing Healthcare Predictive Models Through Privacy- Preserving Synthetic Data Generation
    (Faculty of Engineering, 2025-09-09) Edirisinghe M.M; Gunarathne J H M S M; Wanniarachchi W A A M
    The advancement of healthcare predictive modeling is closely tied to the availability and quality of patient data. However, privacy regulations and ethical concerns often hinder data sharing, making it a persistent challenge. As a solution, privacy-preserving synthetic data generation has emerged, enabling the creation of artificial datasets that retain the statistical properties of real data while protecting individual privacy. This paper explores the use of such synthetic data throughout the clinical risk prediction pipeline by leveraging state-of-the-art generative models. We evaluate their utility in exploration data analysis, feature selection, model training, and deployment. Our study focuses on synthetic data generated using advanced models such as Differentially Private GANs (DPGAN), Private Aggregation of Teacher Ensembles GANs (PATEGAN), and Anonymization through Data Synthesis GANs (ADSGAN). Using these techniques, we created synthetic versions of the UK Biobank ever- smoker cohort. These synthetic datasets were shown to reproduce key statistical patterns, support effective feature selection, and enable accurate lung cancer risk prediction modeling all without using real patient data. We compare synthetic data with other privacy-enhancing technologies like federated learning and highlight a key advantage: synthetic data allows the direct use of existing analytical and machine learning tools without modification. Additionally, we examine deployment models such as "no- release" and "delayed-release," emphasizing how synthetic data can speed up research and enable broader data sharing while maintaining GDPR compliance. Overall, this study demonstrates the potential of synthetic data to transform healthcare research, software testing, education, and collaboration while carefully navigating the trade-off between privacy and utility.
  • Thumbnail Image
    PublicationOpen Access
    A Data-Driven Approach to Predicting Ischemic Heart Disease Risk in Monaragala: Integrating Lifestyle and Symptom Factors with Machine Learning
    (Faculty of Engineering, 2025-09-09) Meddepola, M.A.R.L.; Wickramasinghe, B.M.G.S.T.S.K.
    Ischemic Heart Disease (IHD) remains a leading cause of mortality worldwide and presents a critical challenge in underserved rural areas such as Monaragala, Sri Lanka. Traditional IHD prediction methods predominantly depend on clinical diagnostics like ECGs and blood tests, which are often unavailable or inaccessible in such regions. This study aims to bridge this gap by developing a machine learning-based prediction model that utilizes only lifestyle and symptom-related data, eliminating the need for invasive clinical procedures. A dataset comprising lifestyle habits (e.g., diet, smoking, alcohol use, exercise) and symptom indicators (e.g., chest pain, fatigue, dizziness) was collected via surveys. Feature selection using Logistic Regression identified the top eight most relevant predictors. Five machine learning algorithms, Logistic Regression, K-Nearest Neighbors, Support Vector Machine, Decision Tree, and Random Forest, were trained and evaluated. Among them, the Random Forest model achieved the highest performance with an accuracy of 83.5%, precision of 0.86, recall of 0.78, and F1- score of 0.81, demonstrating strong predictive capability based solely on non-clinical features. In addition, a web-based self-assessment tool was developed to make the model accessible to the public, particularly targeting individuals in rural areas with limited healthcare access. The tool enables users to input basic lifestyle and symptom information and receive a real-time risk assessment. The findings confirm that the model leveraging lifestyle and symptom data can effectively identify individuals at risk of IHD. This approach supports the development of scalable, low-cost, and user-friendly screening tools that can enhance early detection and preventive care, especially in rural and resource-constrained settings.
  • Thumbnail Image
    PublicationOpen Access
    Fitness Warrior: Fitness and Nutrition Tracker with Personalized Goal Generation
    (SLIIT City UNI, 2025-07-08) De Mel, Y.D; Nallapperuma, P.M
    Fitness Warrior is a comprehensive mobile fitness tracking application developed using React Native and Firebase that addresses critical limitations in existing solutions through the innovative integration of machine learning, gamification, and social features. Traditional fitness applications suffer from inaccurate step detection (with error rates exceeding 20% error rates), inefficient nutrition tracking interfaces, poor user retention (with 73% abandonment within three months), and a lack of adaptive personalization. This project uniquely implements ondevice machine learning via TensorFlow.js for privacypreserving step detection, combines TF-IDF vectorization with cosine similarity for efficient food searching, and incorporates principles of Self-Determination Theory through a cohesive social motivation framework. Development followed the Agile Scrum methodology, implementing a CNN-based model processing sensor data at 50Hz sampling rate, creating a database of 2,395 food items with optimized search algorithms, and designing gamified social features. The application achieves 95.2% real-world step counting accuracy compared to manual counting, significantly outperforming conventional threshold-based approaches (48.3% accuracy), while the calorie tracker delivers 92.7% relevant results in top-5 suggestions with 126ms search latency. Evaluation with 21 users demonstrated exceptional impact: 95.3% reported increased daily steps, 90.4% experienced greater calorie intake awareness, and 71.4% found social features strongly motivating. The application received outstanding approval with 90.5% of testers rating overall satisfaction at 8 or higher on a 10-point scale. This research successfully demonstrates how integrated, machine learning-enhanced fitness applications can meaningfully impact user health behaviours while overcoming significant limitations in existing solutions.
  • Thumbnail Image
    PublicationOpen Access
    Smart Chat: A Mobile Chat Application Based on Machine Learning
    (SLIIT City UNI, 2025-07-08) Yathuraj, G; Abeysinghe, A
    In an increasingly digital world, communication is primarily conducted through messaging apps, but these platforms cannot often convey emotional nuance. This limitation can lead to misunderstandings, emotional disconnects, and deteriorating relationships. SmartChat addresses this gap by integrating machine learning-based emotion recognition into a mobile chat app, allowing users to send and receive voice messages enriched with emotional context. Built using React Native and compatible with both Android and iOS, SmartChat analyzes voice cues such as tone, pitch, and cadence to detect and display emotions to the user. This innovation improves the clarity and empathy of conversations, making digital communications more humancentered. Beyond general messaging, SmartChat has the potential to be used in critical contexts such as education, mental health support, and emotional literacy. By making emotionally aware communication accessible across languages and cultures, SmartChat contributes to fostering healthy interpersonal relationships and supports the broader goal of social sustainability through technology.
  • Thumbnail Image
    PublicationOpen Access
    AI-Driven To-Do List: Optimizing Task Categorization and Prioritization Using Ensemble Models
    (SLIIT City UNI, 2025-07-08) Vishaliney, P.; Pemasiri, C.S; Kanthakumar, M; Yatigammana, N
    This paper introduces the AI-driven smart todo list that can cluster and prioritize activities by using the machine-learning methods. Traditionally, to-do list services are immovable and have an element of compromising users to input the information themselves; this sort of bare tool can easily lead to unproductiveness in accomplishment of duties. To address this situation, we supplement ensemble modeling, namely Logistic Regression, XGBoost, and Multilayer Perceptron, to delegate the tasks to the desired categories and define priorities by their urgency. Measured based on standard measures, the ensemble will achieve 47.7 percent accuracy when doing classification and 72.8 percent when predicting priority, and High Priority tasks will gain in this evaluation. Using BERT-based embeddings in combination with TF-IDF-based vectorization, the system should improve its effectiveness because it understands the semantics of described tasks. Together these blocks form a superb ensemble architecture that can beat stand-alone model when it comes to classification and forecasting. More importantly, the system still leaves itself potential to adjust to user behavior and therefore it can improve task management, and it is a feasible platform in real time organization of tasks.