Institutional-Repository, University of Moratuwa
Welcome to the University of Moratuwa Digital Repository, which houses postgraduate theses and dissertations, research articles presented at conferences by faculties and departments, university-published journal articles and research publications authored by academic staff. This online repository stores, preserves and distributes the University's scholarly work. This service allows University members to share their research with a larger audience.
 Research Publications |  Thesis & Dissertation |  E- Books |
 UoM Journal Publications |  UoM Conference Proceedings |  Articles published in Scimago's Q1 journals |  UoM Research Reports |  Other Articles authored by UoM staff |
Recent Submissions
item: Conference-Full-text
Evaluating the quality of traditionally peeled vs. machine peeled cinnamon
(IEEE, 2025) Alahakoonge, AD; Gunasekara, H; Amarasinghe, S; Gamage, J
Ceylon cinnamon (Cinnamomum zeylanicum), which is primarily grown in Sri Lanka, is prized across the world for its unique flavor and aroma. However, due to the increasing cost of labor and a shortage of skilled workers, the traditional hand-peeling method is slow, labor-intensive, and becoming less economically feasible. Technologies for semiautomated peeling are being investigated as a solution to these issues to increase processing effectiveness while maintaining product quality. Key quality parameters such as cinnamon peel thickness, color uniformity, and oleoresin yield will be evaluated in this study in order to compare the efficacy of machine peeling versus traditional methods. Different combinations of the scraping, rubbing, and peeling processes were evaluated using a full factorial experimental design. Neither the scraping, rubbing, or peeling methods individually had a significant effect on thickness or oleoresin yield, according to the study, but the combination of the three processes had a significant effect on peel
thickness. For thickness and oleoresin yield, statistical analysis using one-sample t-tests and multi-way ANOVA demonstrated no significant differences between traditionally and machine-peeled samples (p > 0.05). These findings demonstrate that machine peeling methods can produce quality on par with traditional peeling methods, providing a cost-effective and environmentally friendly alternative. By establishing a quality foundation for machine-peeled cinnamon that satisfies industry standards while maintaining key product qualities, this study contributes to modernizing Sri Lanka’s cinnamon industry.
item: Conference-Full-text
Classification of same limb motor imagery EEG using temporal attention based hierarchical transformer
(IEEE, 2025) Liyanage, M; Gunasekara, P; Laksara, R; Ranaweera, R; Wijayakulasooriya, J; Harischandra, N; Dassanayake, T
Applications of motor imagery (MI) based braincomputer interfaces (BCI) are frequently seen in medicine and robotics. Currently, most BCIs rely on distinct body parts such as the left hand, right hand, feet, and tongue. However, due to the limited number of independent control signals they provide, these are not ideal for complex system control. MI tasks within the same upper limb address this by enabling more intuitive system control, but have relatively few studies on them. It is challenging to classify these tasks because they activate closely spaced motor cortex regions. To address this, we propose an attention mechanism based hierarchical transformer architecture that selectively emphasizes temporal segments with important features by assigning higher attention weights. It consists of a low level transformer (LLT) layer that extracts features from short EEG segments, and a high level transformer (HLT) that uses selfattention to identify and combine key features for classification. The model achieved a competitive accuracy of 54.07% for four class same limb MI tasks, far surpassing the 25% chance level, and also demonstrated a reasonable robustness to session variability. Experimental results indicate the model’s effectiveness in classifying MI tasks of the same limb and potential for the advancement of BCIs.
item: Conference-Full-text
Sinhala language specific vocal biomarker extraction for parkinson’s diagnosis using machine learning
(IEEE, 2025) Rathnayake, PS; Lokuhennadige, LCD; Samarakoon, SMUS; Herath, HMKKMB; Madhusanka, BGDA; Yasakethu, SLP
Early detection of Parkinson's disease (PD) remains challenging, particularly in underrepresented populations where diagnostic resources are limited and culturally appropriate screening methods are needed. Limited research exists on using Sinhala voice biomarkers for PD detection in Sri Lankan populations through machine learning approaches. Speech data were collected from Sri Lankan PD patients and healthy controls, focusing on simple vowel sound articulation. Voice biomarker features were extracted and analyzed using four machine learning classifiers: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest (RF), and Logistic Regression (LR). Models were evaluated using precision, recall, accuracy, and F1-score metrics. A hybrid voting classifier combined the bestperforming algorithms. SVM and LR demonstrated superior performance among individual classifiers. The hybrid voting classifier combining these algorithms achieved 87% accuracy in detecting early-stage PD. Vocal biomarkers and hybrid machine learning strategies show promise for early PD recognition in marginalized populations.
item: Conference-Full-text
GO-PES membrane for industrial dye effluent water purification
(IEEE, 2025) Athapaththu, S; Gamage, N; Sitinamaluwa, H
Clean water scarcity is a major global issue, posing significant challenges for both the environment and human health. A major concern in industrial wastewater management is the presence of elevated concentrations of dyes in water systems. Graphene-based nanomaterial membranes offer a proactive solution, effectively removing industrial dye contaminants from water. The intrinsic two-dimensional structural attributes and remarkable properties exhibited by graphene and grapheme oxide (GO) provide opportunities for their integration into nano-porous materials. When combined, these materials offer modifiable characteristics, enabling fine-tuning for enhanced efficacy in water filtration applications. Utilizing a pressureassisted technique, synthesized GO-PES (GO-Poly ether sulfone) nano porous membranes demonstrate heightened efficacy in the removal of Methylene Blue (MB) and Methyl Orange (MO), excelling particularly in key parameters such as membrane selectivity and permeation flux. In this study, industrial dye filtration membranes were synthesized using four different concentrations of GO to modulate the amount of GO incorporated. The results reveal notable trends, as GO concentration increase, a increment of selectivity and decrease of flux was observed. A maximum selectivity of 72.4% was observed and a maximum flux of 0.03332 m3/m2s was also observed.
item: Conference-Full-text
SinLlama - a large language model for Sinhala
(2025) Aravinda, HWK; Sirajudeen, R; Karunathilake, S; De Silva, N; Kaur, R; Ranathunga, S
Low-resource languages such as Sinhala are often overlooked by open-source Large Language Models (LLMs). Therefore, it is imperative that the existing LLMs are further trained to cover such languages. In this research, we extend an existing multilingual LLM (Llama-3-8B) to get a better coverage for Sinhala. We enhanced the LLM tokenizer with Sinhala specific vocabulary and performed continual pre-training on a 10 million sentence Sinhala corpus, resulting in the SinLlama model. This is the very first decoder-based open-source LLM with explicit Sinhala support. When SinLlama was instruction fine-tuned for three text classification tasks, it outperformed base and instruct variants of Llama-3-8B by a significant margin.