University of Colombo e-Repository

UCER (University of Colombo Electronic Repository) is a collection of scientific research publications by researchers at the University of Colombo, Sri Lanka. This e-Repository serves to manage, preserve and make available the academic works of the faculty, postgraduate students, and research groups. The collection includes faculty publications, master's and doctoral theses abstracts. This repository is updated regularly, and new works are added to collections on a continuous basis

Guidelines

Authors are responsible for obtaining copyright permission from the publisher and submitting the signed declaration to ir@lib.cmb.ac.lk.

Recent Submissions

  • Item type: Item ,
    Prospective Teachers' Awareness and Readiness for Artificial Intelligence Integration in Teaching
    (University of Colombo, 2026) Nawastheen, F. M.; Dilakshan, R. S.; Fernando, W. M. P. A. S.; Weerasighe, M. D. B. P.; Sandamali, K. A. D.
    Teacher education plays a vital role in preparing future teachers to respond to technological developments. Artificial Intelligence (AI) is increasingly influencing teaching and learning, creating opportunities and challenges for teacher preparation. This study aimed to understand prospective teachers’ preparation for AI integration by examining their awareness of AI tools, self-reported readiness to integrate AI into teaching, and perceptions of institutional support. A quantitative survey design was employed with a purposive sample of 98 prospective teachers enrolled in a Bachelor of Education programme at an Open and Distance Learning institution in Sri Lanka. Data were collected using a literature-based structured questionnaire covering demographic information, AI awareness and usage, AI literacy, teacher AI readiness, attitudes and behavioural intention, risk awareness, and open-ended questions. The questionnaire was reviewed by two experts and piloted with 20 students. Cronbach’s alpha coefficients ranged from .768 to .863. Data were analysed using percentages, means and standard deviations. Respondents were highly familiar with general-purpose tools such as ChatGPT and Gemini but had limited awareness of education-specific AI tools. Ethical Teaching Practice received the highest mean rating (M = 3.67, SD = 0.74), while Pedagogical Application (M = 3.35, SD = 0.82) and Self-Efficacy (M = 3.41, SD = 0.61) received comparatively lower ratings. Perceived Institutional Support received the lowest rating (M = 3.04, SD = 0.67). These findings reflect self-reported perceptions rather than demonstrated teaching competence. Teacher education programmes should provide structured AI learning, practical training and continuing institutional support to prepare prospective teachers for effective and ethical AI integration. Future research should include B.Ed. students from different language media and disciplines to examine variations in awareness, readiness and perceived support.
  • Item type: Item ,
    Economic Instability and Educational Participation in Sri Lanka
    (University of Colombo, 2026) Gayathiri, B.; Branavan, S.
    Economic instability has emerged as a major challenge in Sri Lanka, as persistent inflation and economic disruptions continue to affect household welfare and educational participation, particularly among low-income families. This study examines the impact of inflation on educational participation, measured by the gross school enrolment ratio, in Sri Lanka over the period 1990 – 2024. Inflation is the principal indicator of economic instability, with GDP per capita growth and government education expenditure (% of GDP) included as controls. Annual data from the World Bank and the Central Bank of Sri Lanka were analysed using the Autoregressive Distributed Lag (ARDL) bounds testing approach. Unit root tests show that inflation and GDP per capita growth are I(0), while enrolment and government education expenditure are I(1), with none integrated of order two. The bounds test (F = 6.77) exceeds the upper critical bound, confirming a long-run relationship. In the long run, inflation (0.092, p = 0.035) and GDP per capita growth (0.347, p < 0.001) are positively and significantly associated with enrolment, while government education expenditure is insignificant (p = 0.535). The error correction term is negative and significant (−1.037, p = 0.001), indicating that deviations from long-run equilibrium are corrected within about one year. In the short run, no variable is significantly associated with enrolment. Enrolment remained within a narrow band even as inflation reached 49.7% in 2022. These findings suggest that free-education policy has protected aggregate participation during macroeconomic shocks, but stable national enrolment should not be equated with equitable access, because aggregate data cannot reveal effects on low-income households. Policymakers should safeguard real education spending against inflation and adopt household-level monitoring systems to advance inclusive education under Sustainable Development Goal.
  • Item type: Item ,
    FloodSense: An IoT-Based Flood Monitoring, Risk Prediction and Community Early Warning System for the Kalu Ganga Basin, Rathnapura
    (University of Colombo, 2026) Wimalaweera, R. B. D. I.; Lochani, K. A. M.; Wijekoon, W. M. A. I. M.; Samarasinghe, N. R.; Wijayarathna, P. H. S. S.
    Flood disasters in Sri Lanka’s Rathnapura District along the Kalu Ganga Basin cause repeated loss of life and economic damage, while current flood management remains largely reactive, relying on manual observation and delayed response. This paper introduces FloodSense, and integrated IoT-based flood monitoring and early warning system combining real-time environmental sensing, machine learning-based flood prediction, and multi-platform alert dissemination. Prototype sensor nodes using ESP32 Microcontrollers and SX1278 LoRa Radio Modules measure water level, rainfall, temperature, and humidity. A low-power “deep-sleep” technique extended battery runtime to over 60 hours, while RTC Persistent-Memory maintained the Rain-Guage Baseline through power-cycles. The combined dataset comprised 2132 daily station-level observations of water level, rainfall, temperature, and humidity from Ellagawa, Putupaula, and Rathnapura, spanning January 2024 to December 2025. Each station was split chronologically 80/20 for training and testing to prevent temporal leakage, while validation monitoring was used during LSTM training. SMOTE was applied only to training data, reducing imbalance while preserving the independent test partition. The two-stage machine-learning pipeline uses an XGBoost Regression Model to predict future water levels (R2 = 0.9476 & RMSE = 0.5588 m), followed by an LSTM Classifier for four severity levels: Normal, Alert, Minor, and Major Flood (Accuracy = 94.80%, Macro-F1-Score = 0.6663). Alternative XGBoost and Random Forest classifiers achieved 100% training and test accuracy but were rejected because serve class imbalance indicated overfitting and poor generalization. Results are displayed through a web-based portal with interactive maps and charts, while public warnings are disseminated through a mobile application using Firebase Cloud Messaging Push Notifications with offline capability during network disruptions.
  • Item type: Item ,
    Cinematic Image and Ideological Representation: A Marxist Reading of Class Consciousness in the Contemporary Sri Lankan Digital Media Streaming Landscape
    (University of Colombo, 2026) Ranathunga, R. M. N. A.
    The contemporary transition of Sri Lankan cinema from traditional theatrical exhibition to localized OTT applications, independent YouTube channels, and global streaming platforms fundamentally alters the visual construction of class dynamics. Addressing the conference theme of advancing an inclusive cultural future, this qualitative study critically investigates how digital media innovation impacts the representation and cultural visibility of economically marginalized communities in post-crisis Sri Lanka. Grounded in Marxist cultural theory specifically Louis Althusser’s Ideological State Apparatuses and Antonio Gramsci’s cultural hegemony the research deconstructs the structural shifts in subaltern representation beyond traditional cinematic boundaries. The methodology employs Christian Metz’s film semiotics to perform textual and visual analysis on a purposively selected corpus of eight contemporary Sri Lankan audio-visual texts (five independent short films and three web series) released on digital platforms between 2022 and 2026. Findings reveal a dual ideological mechanism: while lowered digital production barriers initially democratize content creation and provide space for subaltern narratives, this inclusivity remains largely superficial. When processed through digital streaming ecosystems, radical working-class consciousness undergoes systemic commodification, where socio-economic struggles are aestheticized into digestible melodrama to align with middle-class digital consumption patterns. Consequently, the study demonstrates that digital platform structures do not automatically liberate the cinematic image; instead, they shift the battlefield of cultural hegemony into the digital domain, neutralizing social discontent into marketable commodities. By mapping this intersection of film semiotics, platform capitalism, and inclusive representation, this study bridges a critical gap in Sri Lankan media scholarship, offering insights into equitable media access in the digital era.
  • Item type: Item ,
    Generative AI-Assisted Short-Form Documentary Storytelling for Preserving Sinhala Folk Tales Among Urban Gen-Z Audiences in Sri Lanka
    (University of Colombo, 2026) Paranamana, P. S.
    The rapid integration of generative artificial intelligence (AI) into creative industries offers new possibilities for digital storytelling and cultural heritage preservation. However, limited research has examined its application to short-form documentary storytelling for safeguarding intangible cultural heritage, particularly Sinhala folk tales in Sri Lanka. This study explores how generative AI-assisted short-form documentary storytelling can contribute to preserving Sinhala folk tales and enhancing audience engagement among urban Generation Z viewers. Guided by digital storytelling theory and the UNESCO framework for safeguarding intangible cultural heritage, it examines AI-assisted production, cultural representation, and audience interpretation.A qualitative, multi-method approach combines semi-structured interviews, practice-based documentary production, and audience focus group discussions. Purposive sampling will recruit 15 participants: five emerging filmmakers, five media practitioners, and five film students with relevant knowledge or experience in AI-assisted media production. Additionally, 18–24 urban Gen Z participants aged 18–29, residing in selected urban areas of Sri Lanka, will participate in three focus groups. ChatGPT will support script generation and narrative structuring, while Midjourney will facilitate visual conceptualisation and reference development for documentary prototypes. Prompts and human editorial decisions will be documented to support transparency, cultural accuracy, and contextual relevance. Focus groups will explore audience engagement, cultural understanding, and interpretive responses. Interview and focus group data will be analysed using thematic analysis. Distinguishing AI-assisted documentary storytelling from conventional digital storytelling and fully AI-generated animation, this study investigates generative AI as a creative support tool rather than a replacement for human authorship and cultural knowledge. It also addresses authenticity, authorship, and cultural representation, contributing to discussions on AI-driven creative practices and innovative approaches to safeguarding intangible cultural heritage in Sri Lanka.