ComURS2026 Computing Undergraduate Research Symposium : Abstracts

Exploring Financial Literacy Gaps and Business Analysis-Driven Strategies for Enhancing User Engagement in PFM Apps in Sri Lanka

2026-06-15T08:25:39Z

Exploring Financial Literacy Gaps and Business Analysis-Driven Strategies for Enhancing User Engagement in PFM Apps in Sri Lanka Nissanka, N.P.S.N.; Herath, G.A.C.A. Personal Finance Management (PFM) applications are being marketed all over the world to enhance budgeting, savings discipline, and financial inclusion. Nevertheless, the penetration of advanced PFM features in Sri Lanka remains low despite the high ownership of smartphones. This paper analyses the major impediments affecting PFM app adoption (Dependent Variable) and offers Business Analysis-based solutions to facilitate greater adoption. A web-based questionnaire was created in a stratified fashion (n = 421). Six theoretically based constructs were measured: Low Financial Literacy, Low Feature Awareness, Low Usability, Lack of Trust, Cultural and Behavioral Habits, and Limited Support and Guidance. The reliability results indicated good internal consistency (Cronbach’s α =0.760 to 0.941). Principal Axis Factoring (Exploratory Factor Analysis; Promax rotation) results showed that sampling adequacy was reached (KMO = .888; Bartlett p < .001) and yielded five factors that explained 60.3 percent of variance. Usability was interrelated with guidance, language, interface clarity, and support-related items, indicating that users perceive direction, language, and interface clarity as one construct. The final model was selected as containing five empirically validated predictors. The multiple regression analysis indicated that the predictors explained 74 percent of the PFM app adoption variance (R2 = .74). The most important negative predictors were trust and security issues, along with usability problems. Cultural propensities towards cash and passbook-related financial behaviors hindered usage, but greater awareness towards features enhanced the chances of using PFM. The paper suggests an integrated approach that would include simplified and localized interfaces, onboarding, culturally sensitive features including community savings tracking and open-line security communication. Business Analysis strategies, such as stakeholder mapping, user-story development and repetitive usability testing are hypothesized to translate these findings into applicable developer, financial, and policy improvements.

Work-life balance and job satisfaction among female IT professionals in Sri Lanka’s technology sector

2026-06-15T08:19:44Z

Work-life balance and job satisfaction among female IT professionals in Sri Lanka’s technology sector Morais, M.M.P.; Ishanka, U.A.P. The number of women in the technology field in Sri Lanka is on the rise, but women in technical jobs complain of struggling to have a balance of work and life (WLB) as they strive to meet the demands of high performance and household women. This paper will look at the predictability of perceived WLB on job satisfaction among female IT professionals in Sri Lanka in response to the paucity of local evidence that is specific to women in technical roles. A cross-sectional survey, conducted online was quantitative and involved 115 female IT professionals working in various fields. WLB and job satisfaction were measured by validated Likert-scale questions. Due to the composite scores, simple linear regression (job satisfaction as the dependent variable) were to be used in order to analyze them. The correlation between WLB and the satisfaction with the job was moderate and positive (r = 0.512, p < 0.001). In regression analysis, the results show that WLB is a significant predictor of job satisfaction (b1 = 0.744, SE = 0.118, t =6.335, p<0.001) with 26.2% of the variance (R2 =0.262, Adjusted R2 =0.256). Results of this sample indicate that the enhancement of the working expectations, and conducive flexibility can become the source of improved job satisfaction of women in Sri Lankan IT environments. To illustrate this point, a simple predictive model (linear regression) was fitted to learn job satisfaction based on WLB; performance on the test was moderate (R2 = 0.18), suggesting that more organizational and personal predictors would probably be required.

Cross-Modal Predictive Modeling of Mental Health Treatment Outcomes: A Machine Learning Framework for Comparing Psychiatric Counseling Therapy and Therapeutic AI-Chatbots

2026-06-10T04:50:35Z

Cross-Modal Predictive Modeling of Mental Health Treatment Outcomes: A Machine Learning Framework for Comparing Psychiatric Counseling Therapy and Therapeutic AI-Chatbots DeSilva, M.T.D.; Kaushalya, P.K.D.K. Mental health problems are becoming the order of the day and burdening the traditional psychiatric guidance frameworks in terms of expenses, unreachability and waiting durations. Conversely, mental health chatbots that are based on AI have become popular because of their anonymity, 24/7, and low cost. Although both traditional counseling and chatbot approaches have feasible advantages, no standard way of operating has been established to compare the efficacy of the two with individual patients. This has been a barrier to the use of individualized mental health interventions. The paper examines the application of machine learning and in this case, the Random Forest algorithm to predict and compare the results of conventional therapy and AI chatbot assistance. Available references define the major signs of treatment success and provide an overview of the benefits and shortcomings of chatbot interventions, yet no model exists to evaluate how people can react to the alternative medium. To solve this, a random forest model was created using data on clinical therapy outcome and the results were used on the data of chatbot users to forecast possible outcomes. The reported chatbot outcomes were compared statistically and through the qualitative feedback with the expected therapy outcomes. The research will establish personal characteristics that relate to increased benefits in either form of therapy. The expected outcomes will be used in clinical decision making, enhancement of digital mental health tools and help in choosing the most appropriate treatment.

Deepfake Image and Video Detection System for Sri Lankan Facial Features Using Machine Learning

2026-06-10T04:44:13Z

Deepfake Image and Video Detection System for Sri Lankan Facial Features Using Machine Learning Karunarathna, A.M.T.H.; Abeythunga, W.M.L.S. Deepfakes and other AI-generated manipulated images and videos have become an increasing cyber threat to Sri Lanka as AI-generated multimedia content becomes more accessible to consumers. Global AI-generated multimedia datasets used in global deepfake detection models do not include sufficient representation of Sri Lankan characteristics including: darker/mixed brown skin tone; South Asian facial structure; ethnic diversity (Sinhalese, Tamil, Muslim, Burgher); traditional clothing; and lighting found in a variety of local environments. Therefore, many international deep fake detection systems are either fail to accurately identify manipulated images of Sri Lankan faces, or fail when detecting low resolution video content captured on mobile devices that are commonly used in Sri Lanka. A system designed to detect deep faked images and videos of Sri Lankan faces is presented in this research. The system uses a CNN-based image forensic model in combination with frequency domain-based artifact analysis and landmark consistency checks to evaluate each image submitted by a user. Additionally, the system also analyzes video submissions for deep fakes by extracting frames from the input video and evaluating each frame individually using the trained image model. Finally, the results from each individual frame are aggregated into an overall decision regarding the authenticity of the video submission. A custom dataset was developed for the purposes of training the system’s models, which focuses on a variety of aspects of Sri Lankan skin tones, facial structures, cultural elements, and environmental factors.