Browsing by Author "Rathnayake, S"

Now showing 1 - 17 of 17

Embargo
Adaptive AI-Based Enhancement of Critical External Sounds in Insulated Vehicle Cabins for Improved Safety
(Institute of Electrical and Electronics Engineers Inc., 2025-12-09) Rathnayaka D.B; Wickramasuriya L.H.N.Y; Walpalage J.V; Rathnayake, S
The increasing acoustic insulation in modern and electric vehicles improves passenger comfort but unintentionally suppresses critical external sounds such as ambulance sirens, car horns, and train alarms, creating potential safety risks. While existing research has explored sound detection or localization in isolation, few systems integrate both capabilities in a unified framework for real-time vehicular deployment. This research proposes an adaptive AI-based system that detects, classifies, and selectively enhances these critical sounds in real time while providing directional awareness. Using a convolutional recurrent neural network (CRNN) trained on the UrbanSound8K dataset, the system processes incoming audio from external microphones, extracts Mel-frequency cepstral coefficients (MFCCs), and distinguishes safety-relevant cues from non-essential background noise. A dual-microphone setup enables the estimation of sound direction (left or right), providing additional spatial awareness to the driver. Detected signals are isolated through spectral filtering and relayed into the cabin with sub-30 ms latency, ensuring timely driver and passenger awareness without compromising comfort. Experimental results achieved 91.2% classification accuracy and 87.4% directional accuracy,confirming the system's feasibility for enhancing safety in insulated vehicle cabins and supporting future autonomous driving environments.
Embargo
Adaptive Video Game Content Generation through Player Centered Modeling
(Institute of Electrical and Electronics Engineers Inc., 2025-12-09) Hapuarachchi H.A.R.S; Herath H.M.N.R; Kaveesha B.G.S; Deheragoda D.M.L.M; Rathnayake, S; Chamara, D
This research introduces a novel, integrated AI system for Adaptive Video Game Content Generation Through Player Centered Modeling, designed to overcome the limitations of conventional static game mechanics by dynamically modifying gameplay elements (levels, quests, music, and enemy behavior) in real-time based on player biometric and behavioral data. Key contributions include the Personalized Quest Generation System, where the CatBoost model performed well in predicting player engagement, and shifting to the YOLO11X-CLS classification model substantially reduced computational lag for real-time emotional adaptation. For Level Generation, the implementation of a bootstrapping methodology during DCGAN training enabled continued refinement, resulting in lower Symmetry Error and Block Diversity Error, while the Intelligent Enemy Agent, trained using Dueling DQN, demonstrated a clear upward trend in Mean Rewards, successfully generalizing learned pursuit strategies to a real-time environment. Despite these successes, limitations include potential mild overfitting in the emotion recognition model, evidenced by a slight increase in validation loss after 20 epochs, and the CNN used for dynamic music adjustment struggled with low-resolution webcam inputs, leading to occasional minor misclassifications and slight delays in music transitions when multiple biometric parameters changed simultaneously.
Embargo
Adaptive Voice Communication in Emotion-Aware Digital Companions
(Institute of Electrical and Electronics Engineers Inc., 2025) Rathnayake, P; Rathnaweera, C; Jithma, U; Aththanayake, I; Rathnayake, S; Gunaratne, M
This paper presents an adaptive voice communication system for emotion-aware digital companions that dynamically responds to users' affective states through expressive speech and synchronized 3D avatar animation. The system integrates real-time voice input, emotion recognition, and context-aware dialogue generation using GPT-3.5, followed by emotional text-to-speech synthesis via neural TTS. Lip-sync data is generated using phoneme alignment and rendered in sync with the avatar's facial expressions and gestures. To enhance user trust and engagement, the avatar visually mirrors the emotional tone of the speech. A cultural adaptation layer is introduced to align voice output and speech style with Sri Lankan communication norms, including tone, pacing, and formality. Implemented using a Node.js backend and React + Three.js frontend, the system demonstrates strong potential for emotionally intelligent, culturally adaptive AI interactions. This work contributes a modular pipeline for building empathetic voice agents capable of enhancing realism and trust in human-AI communication.
Open Access
Determining the influence of LPI, GCI and IR on FDI: A study on the Asia and Pacific Region
(PLoS ONE, 2023-02) Wannisinghe, P; Jayakody, S; Rathnayake, S; Wijayasinghe, D; Jayathilaka, R; Madhavika, N
Competitiveness Index (GCI) and Interest Rates (IR) on Foreign Direct Investment (FDI) for the Asia & Pacific region. The study is original as extensive evidence on the impact of LPI, GCI and IR on FDI in the Asia & Pacific region are examined initially. For the years 2007, 2010, 2012, 2014, 2016 and 2018, data was gathered for 33 nations in the Asia and Pacific area. Data analysis was performed using a panel regression model and multiple linear regression. The findings of the study reveal that LPI, GCI and IR are the three major factors influencing FDI inflows into the economies. However, the impact of these factors varies from country to country. The results concluded that LPI positively impacts FDI in India, Korea, Lebanon, and Oman. In contrast, a negative influence was observed for China, Kuwait and the Philippines. GCI positively impacts FDI in China, Korea, Kuwait, Pakistan and the Philippines, while a negative impact was observed in Armenia, India, Lebanon. Furthermore, IR has a positive impact on FDI flows in China and Egypt while in Korea and Lebanon, a negative impact was observed. Therefore, policymakers should focus more on improving the infrastructural requirements and macroeconomic factors while considering the other country-level variables that influence the FDI in flow
Open Access
Disease Identification and Mapping using CNN in Paddy Fields
(Faculty of Humanities and Sciences, SLIIT, 2023-11-01) Sandeepanie, W.D.N; Rathnayake, S; Gunasinghe, A
Rice, a globally vital staple crop, sustains over half of the world’s caloric needs while supporting the livelihoods of small-scale farmers and landless laborers. The escalating global population has led to an increased demand for rice production. Sri Lanka, renowned for its premium rice quality, has a rich history of paddy cultivation. However, a substantial portion of the country’s 708,000 hectares of paddy land remains underutilized due to water scarcity and unstable terrain. The objective of this project is to enhance paddy crop quality during the critical vegetative phase by employing machine learning and web development for early disease identification. The vegetative phase significantly influences overall yield, resistance to pests and diseases, nutrient assimilation, and environmental sustainability in agriculture. This project primarily focuses on early disease identification during this phase and presents the findings through a user-friendly map interface. Early identification of paddy diseases is vital for effective crop management and high yields. These diseases, caused by various pathogens, can severely impede plant growth and productivity if not promptly detected and treated. Identifying them early enables farmers and experts to take timely, targeted actions such as applying suitable fungicides or implementing cultural practices to control their spread and minimize crop damage. A logical map, displaying disease spread percentages, will gauge the impact of infections on paddy plants. The reliability of this mapping process hinges on model accuracy, which was rigorously validated using multiple metrics to ensure its effectiveness.
Open Access
From Off-time to Offline: Recognising Right to Disconnect under the Sri Lankan Labour Law System
(Faculty of Humanities and Sciences, SLIIT, 2023-11-01) Rathnayake, S
This study aims to interrogate the possibility of acknowledging the Right to Disconnect (RTD) under the labour laws of Sri Lanka. RTD which emerged as a response to the increasing demand for online and hybrid work, recognises employees’ right to remain disconnected or disengaged from any work-related activities such as answering phone calls or responding to emails after working hours. Research on RTD has focused on its practical implications for building a healthy and sustainable work force, but less has been said about its incorporation to domestic legal environments, especially in developing countries. The present study therefore critically reviews the Sri Lankan labour laws and the labour laws in other jurisdictions that have acknowledged RTD, to understand the potential of acknowledging this right in Sri Lanka. Through its data analysis this study revealed that, despite the existence of legal measures that ensure employees’ Right to relax and leisure (RRL), a little to no focus has been given to the new working environments such as online and hybrid work, and the rights that would emerge from them. While highlighting the best practices relating to RTD from other jurisdictions, this study also examines the legal dilemmas that may thwart such attempts of recognising this right in Sri Lanka. Thus, it highlights the need to update Sri Lankan labour laws to match the contemporary labour market while strengthening labour tribunals to play an active role which will be essential to acknowledge novel rights such as RTD under Sri Lankan labour laws.
Embargo
HCLIP: Beyond CLIP for Cost-Effective Multimodal Retrieval in Education
(Institute of Electrical and Electronics Engineers Inc., 2025-12-09) Weerasinghe, S; Gunatunga, O; Dewpura, W; Fernando, S; Kasthurirathna, D; Rathnayake, S
Multimodal retrieval systems have gained significant attention due to their ability to process and cross-retrieve data containing images and text. However, the factors such as high cost of development, limitation on resources, and the proper addressing of the modality gap, the inherent representational differences between modalities pose a challenge to building effective and efficient retrieval models. In this work, we propose a low-resource, cost-efficient hybrid multimodal retrieval model that integrates Contrastive Language-Image Pre-training (CLIP) and All-MiniLM-L6-v2 to create a shared embedding space while storing raw images in an unstructured database. Our primary contributions include (1) the development of a hybrid model that outperforms CLIP-native retrieval, (2) a novel bidirectional neural network alignment technique that brings textual and visual modalities closer together, and (3) a comprehensive analysis of the modality gap's impact on downstream retrieval performance. Through proper evaluation using transparent techniques such as Mean Reciprocal Rank (MRR) and Cosine-Weighted MRR, our method demonstrates improved retrieval accuracy over baseline approaches. Experimental results exhibit that a lower modality gap does not always prove to be efficient on the downstream retrieval. Our findings pave the way for more efficient, adaptable, and cost-effective multimodal retrieval methodologies in low-resource environments, not limited to the education domain.
Embargo
Human-Ai Cooperative Driving Through Emotion-Aware Decision Making and Driver Personalization
(Institute of Electrical and Electronics Engineers, 2025-12-09) Wickramasuriya L.H.N.Y; Rathnayaka D.B; Walpalage J.V.; Rathnayake, S
Conventional in-vehicle safety systems often neglect real-time emotional monitoring, prediction, and passenger influence, leading to reactive rather than proactive interventions. This paper presents an emotion-aware cooperative driving system that combines facial emotion recognition, time-series emotion forecasting, and personalized music-based regulation. The system detects both driver and front-seat passenger emotions through a Vision Transformer (ViT) model, while a time-series model anticipates the driver's upcoming emotional state. A prioritization algorithm ensures driver emotions hold precedence, with passenger states considered when the driver is stable. Based on this prioritized emotional context, the system regulates the in-cabin atmosphere using music recommendations drawn from the driver's preferred artists via the Spotify API. Experimental results show robust real-time emotion classification (86.4% validation accuracy), proactive forecasting (81.6% predictive accuracy), and improved driver acceptance due to personalization. The proposed framework advances intelligent transportation by shifting from static monitoring toward predictive, human-centered, and non-intrusive emotional regulation, thereby enhancing both safety and user comfort.
Embargo
Hybrid Motion Prediction for Autonomous Vehicles using GNN-Transformer Architecture
(Institute of Electrical and Electronics Engineers Inc., 2025) Akalanka, A; Athukorala, D; Ganepola, N; Tharindu, I; Rathnayake, S
Accurate perception and scene understanding are pivotal in enabling autonomous vehicles to navigate safely and intelligently. This paper presents an integrated perception module comprising three core subcomponents: real-time object detection using YOLOv5, lane-keeping using a CNN-based steering predictor, and a novel motion prediction architecture based on a hybrid Graph Neural Network (GNN) and Transformer design. The system is deployed and validated within the CARLA simulation environment, with custom data generation pipelines designed to mimic real-world behavioral patterns of nearby agents. The novelty lies in the hybrid GNN-Transformer model, which effectively captures both spatial and temporal interactions of dynamic objects for behavior classification. Experimental results demonstrate a high accuracy of 98.75% in classifying behaviors into four categories: Going, Coming, Crossing, and Stopped. This paper details the architecture, dataset creation, training methodology, and performance evaluation, highlighting the hybrid model's potential to improve trajectory planning modules in autonomous systems.
Open Access
In Silico Identification of Abiotic Stress-Responsive Candidate Genes in Rice and Their Allelic Differences Between At 354 and Bg 352
(Faculty of Humanities and Sciences, SLIIT, 2023-11-01) Rathnayake, S; Kottearachchi, N; Lankika, S. P. C.
Rice (Oryza sativa L) feeds more than half of the world’s population. One of the main elements that harm yield globally is abiotic stress. Therefore, it is important to develop abiotic stress tolerant rice varieties in order to increase rice productivity and to extend the cultivation. The lack of knowledge of the genetic mechanisms underlying abiotic stress tolerance is the primary issue with the traditional breeding technique. Hence, studying genes responsible for abiotic stress mechanisms is important to accelerate breeding by molecular marker - based detection techniques. Aiming at finding the candidate genes for abiotic stress tolerance ,two rice genome sequences of At 354 and Bg 352 varieties given by National Research Council, Sri Lanka -16 -16 project were analyzed. At 354 has some abiotic stress tolerance (salt) traits and Bg 352 has some susceptible traits. Next-generation sequencing-derived genome sequences were used to identify SNPs and Indels in the At 354 and Bg 352 varieties with reference to Oryza sativa japonica group cultivar Nipponbare. The STRING Database was used to extract the most correlated genes with abiotic stress. The allelic differences among Nipponbare, At 354 and Bg 352 sequences were detected from Multiple Sequence Alignment by using the Rice Annotation Project database, UGENE software and MEGA 11 software. The mutations of the genes were validated if they were present in another germplasm in the NCBI database. Altogether 100 genes were used to examine, and 166 mutations were observed including 163 SNPs and 3 Indels while 09 genes were validated due to their presence in other rice accessions. The amino acid sequences of the validated sequences were determined by Expasy Translate tool. The Swiss model database and ProtParam tool were used to predict the protein structures and their parameters, which showed some structural differences among tested alleles. These mutant alleles further need to be assessed against abiotic stress and varietal turnover in order to use them in rice improving breeding programs to be used in abiotic stress-prone ecosystems.
Embargo
Knowledge Graph-Based AI Framework for Predicting Nutritional and Health Impacts of Food Ingredients
(Institute of Electrical and Electronics Engineers Inc., 2026-08-04) Dakshina P.D.S.D; Rupasighe W.A.R.K; Waduge N.P; Nimsitha M.V.T; Tissera, W; Rathnayake, S; Krishara, J
The increasing complexity of modern food products and dietary supplements has made it challenging for both consumers and healthcare professionals to interpret nutritional information and assess the potential health risks associated with these products. Modern food labeling schemes provide static and fragmented information and cannot effectively capture the relationships between different ingredients, nutrients and their health effects. In this study, a new AI-based framework named Food Health Risk Analyzer has been proposed that utilizes KGs, GNNs, RAG and a dose-response module based on consumption quantities to perform the dynamic, explainable and evidence-based prediction of food-related health risks. The model uses heterogeneous data in order to analyze the relationships between ingredients and diseases to predict potential health risks while generating scientifically supported explanations as well. The experimental evaluation has shown high prediction accuracy with a micro-F1 score of 0.88 and AUC of 0.85 which shows that the framework surpasses conventional machine learning baseline models. In addition to that, the use of RAG has helped in improving the interpretability of predictions through evidence-based natural language explanations whereas dose-response module improves the practical relevance of risk assessment by considering the consumption quantities of ingredients.
Open Access
Macroeconomic factors affecting FDI in the African region
(PLOS ONE, 2023-01-23) Rathnayake, S; Jayakody, S; Wannisinghe, p; Wijayasinghe, D; Jayathilaka, R; Madhavika, N
Foreign Direct Investment (FDI) occurs when one country invests in another. Multiple factors have contributed to fluctuations in FDI flows globally. This study investigates the impact of the Logistics Performance Index (LPI), Global Competitiveness Index (GCI) and Interest Rates (IR) on FDI in the African region. The study is significant because the African region is underdeveloped and with an unstable macroeconomic environment. Data were collected for 26 countries in the African region for the years 2007, 2010, 2012, 2014, 2016 and 2018 and analysed using Panel Regression and Multiple Linear Regression models. The study’s findings concluded that LPI, GCI, and IR are three major macroeconomic factors impacting FDI inflows. The results indicated that LPI positively impacts FDI in Gambia, Lesotho and Rwanda, while in contrast, LPI impacts FDI negatively in Mauritius. GCI has a positive impact on FDI in Algeria and Lesotho with a negative impact in Rwanda, Mauritius and Namibia. Moreover, IR has a negative impact on FDI in Algeria, Rwanda and Mauritius with a positive impact in Lesotho. Policymakers should pay more attention to the infrastructure development and management of macroeconomic and other factors affecting FDI.
Embargo
Multimodal Knowledge Graph for Domain-Specific Intelligence
(Institute of Electrical and Electronics Engineers Inc., 2025-06-25) Mohan, K; Munasinghe, M; Bandara, L; Wijesinghe, H; Rathnayake, S; Abeywardhana, L
In the era of information abundance, transforming vast amounts of data into meaningful knowledge remains a critical challenge, especially in domains like medicine, engineering, and education, where visual and multimodal elements play a vital role. Traditional Knowledge Graphs (KGs) excel in organizing structured and textual data but struggle to incorporate multimodal information and implicit relationships, limiting their effectiveness. This paper explores the potential of Multimodal Knowledge Graphs (MMKGs) to address these limitations by integrating text, images, videos, and audio into a unified framework. We investigate how MMKGs enhance knowledge retrieval, comprehension, and interactive learning through advanced techniques, including Natural Language Processing and deep learning. Our findings demonstrate that MMKGs significantly improve knowledge retention and application in specialized fields, offering a foundation for more intuitive and effective domain-specific knowledge ecosystems.
Embargo
Multimodal Knowledge Graph for Domain-Specific Intelligence
(Institute of Electrical and Electronics Engineers Inc., 2025) Mohan, K; Munasinghe, M; Bandara, L; Wijesinghe, H; Rathnayake, S; Abeywardhana, L
In the era of information abundance, transforming vast amounts of data into meaningful knowledge remains a critical challenge, especially in domains like medicine, engineering, and education, where visual and multimodal elements play a vital role. Traditional Knowledge Graphs (KGs) excel in organizing structured and textual data but struggle to incorporate multimodal information and implicit relationships, limiting their effectiveness. This paper explores the potential of Multimodal Knowledge Graphs (MMKGs) to address these limitations by integrating text, images, videos, and audio into a unified framework. We investigate how MMKGs enhance knowledge retrieval, comprehension, and interactive learning through advanced techniques, including Natural Language Processing and deep learning. Our findings demonstrate that MMKGs significantly improve knowledge retention and application in specialized fields, offering a foundation for more intuitive and effective domain-specific knowledge ecosystems.
Embargo
NSCLC 360 - Leveraging Multi-Omics Data for a Holistic and Explainable Decision Support for Non Small Cell Lung Cancer Management
(Institute of Electrical and Electronics Engineers Inc., 2025-07-03) Pirabaharan, A; Irfan, A. A.; Lareef, W; Ahamed, S; Rathnayake, S; Shyamalee, T
Non-Small Cell Lung Cancer (NSCLC) remains a leading cause of cancer-related mortality, with existing diagnostic and prognostic models often failing to capture the complexity of tumor biology. This study proposes a holistic and explainable decision support system that integrates multi-omics data - including genomics, transcriptomics, and proteomics - along with advanced machine learning (ML) and deep learning (DL) techniques to enhance NSCLC detection, prognosis prediction, complication forecasting, and recurrence assessment. To address the challenge of interpretability in AI-driven healthcare, we incorporate Explainable AI (XAI) methods such as SHAP and LIME, ensuring model transparency and clinical trust. Additionally, traditional statistical models like Cox proportional hazards regression are combined with ML approaches for robust survival analysis, while modern AI architectures, including Vision Transformers and multi-task learning models, improve tumor localization and TNM classification. By developing an interpretable and clinically meaningful AI-based decision support system, this research aims to advance personalized lung cancer management and improve patient outcomes through seamless integration into clinical workflows.
Embargo
Object Detection Approach for Pure and Cross Chicken Breed Identification
(Institute of Electrical and Electronics Engineers Inc., 2025-12-09) Jayarathna, N; Rathnayake, S; Panduwawala, P
The proposed system aims to identify different types of purebred and crossbred chicken breeds across ten categories. Traditional methods such as visual inspection are often subjective and inaccurate, making breed identification challenging. To address this, image processing and deep learning techniques were employed, with the YOLOv5 object detection algorithm trained on a custom data set of 1,310 images. The model achieved strong results, with an overall precision of 83.5%, recall of 81.8%, and mAP@0.5 of 89.3%. The Class-wise evaluation showed particularly high performance for the Brahma and Leghorn breeds. Based on these outcomes, a mobile application was designed to provide farmers with a fast, reliable, and cost-effective tool for breed identification. In addition to classification, the application provides detailed information on breed characteristics and commercial value, helping small-scale farmers improve productivity, efficiency, and animal welfare
Open Access
Paddy Disease Identification and Impact Calculation Using Machine Learning
(SLIIT Business School, 2023-12-14) Sandeepanie, N; Rathnayake, S; Gunasinghe, A
Rice is a crucial staple crop globally, providing over half of humanity's caloric intake. It supports the livelihoods of small-scale farmers and landless laborers worldwide. With the growing population, there is a high demand for rice production. Sri Lanka is renowned for its high- quality rice and has a long history of paddy cultivation. However, not all the country's 708,000 hectares of land dedicated to paddy cultivation are utilized due to water scarcity and unstable terrain. The objective of this paper is to explore the ways of enhancing the quality of the paddy crop during its vegetative phase by early identification of diseases through the utilization of emerging technologies. The vegetative phase constitutes a critical stage in the growth of paddy, exerting significant influence on the overall yield, resistance to pests and diseases, nutrient assimilation, and the environmental implications of agricultural practices. The primary emphasis of this paper is to identify diseases to which paddy crops are susceptible during the vegetative phase and subsequently present avisual representation of their locations on a map, serving as the output for end-users. Early identification of paddy diseases is crucial for effective crop management and high yields. These diseases, caused by different pathogens, can significantly hinder plant growth and productivity if not detected and treated promptly. Identifying them early allows farmers and experts to take timely and targeted actions, like applying suitable fungicides or implementing cultural practices, to control their spread and minimize crop damage.