Browsing by Author "Dissanayake, D. M. C."

Now showing 1 - 2 of 2

The COVID 19 Pandemic: Challenges Related Online Education on Students of Rural Areas in Sri Lanka
(Faculty of Graduate Studies - University of Kelaniya, Sri Lanka, 2021) Dissanayake, D. M. C.
The recent epidemic of COVID-19 is challenging for each sector like economy, education, health, industries and education. In the field of education sector of Sri Lanka have implemented the online teaching and learning programs for students to minimize the academic loss in distance education. This pandemic has imposed enormous challenges on traditional education process of Sri Lanka. As a consequence of the unprecedented global pandemic, online education has become a new concept and it has created a discrimination between rural and urban students in Sri Lanka. In this background this study examined the significant challenges related online education on students of rural areas in Sri Lanka during COVID-19. Specific objectives are to identify teachers’ attitudes of online education on students of rural areas and the delivery service system and policies related online education in Sri Lanka. This research was used Partially mixed concurrent qualitative dominant status under the qualitative dominant mixed method approach and conducted in Kotavehera educational division, Kurunegala district. Data were collected from 50 students through stratify sampling method and 8 teachers, 5 parents and 2 principals used as key informants. Questionnaires and semi-structured interviews were used as primary data collection technic through telephone conversations. According to the findings of the study it has revealed that lack of availability of low-quality internet, lack of devices, shortage of devices, outdated devices were caused to decrease interest in the subject of students in study area, lack of physical and human resource, negative attitudes, lack of practical knowledge and experience were challenged for create fear and anxiety using of English language and technology in online education of students of rural areas, Parents’ low socio-economic situation were prohibited online education of students and this situation have created troublesome and lot of pressure for students, parents’ personal beliefs and misconceptions, lack of technological skills and teaching methodologies of teachers were directly highlighted as challenges for online education of students in rural areas, Overall, the current curriculum contents of education system in Sri Lanka mainly based on traditional system and it is show lack of flexibility and was not modern and innovative and online teaching methods and abilities are not familiar for students and teachers. There are need to revise the curriculum contents to eliminate the discrimination between the rural urban students and face the sudden situations and education policy makers should attention for revise and update the education curriculum according to the global changes in education. Further ministry of education in Sri Lanka should implement necessary training programs to engage in online education within the traditional system, must give a closer attention to eliminate the disparities in the human and physical resource between rural and urban schools and the government of Sri Lanka with the other non-government agencies should organize the project to supply technological devices for students of rural areas.
QUBO formulation of the closest string problem
(Faculty of Science, University of Kelaniya Sri Lanka, 2023) Dissanayake, D. M. C.
The closest string problem is an NP-complete problem which appears more commonly in bioinformatics and coding theory. From a theorist’s point of view, more interesting applications are found in automata theory, specifically in the context of Finite Automata and Push-Down Automata. The corresponding optimisation problem deals with determining the string corresponding to the minimal Hamming distance between the candidate string and all given strings. Less surprisingly, classical approaches have been pursued with two prominent algorithms being the genetic algorithm and simulated annealing. Latest improvements to quantum computing devices with a specialisation in optimisation tasks, such as D-Wave systems, suggest that an attempt to embed the problem in a model accepted by such systems is worthwhile. In this study, the interests lie in modelling the problem based on binary variables. More specifically, two slightly different Quadratic Unconstrained Binary Optimization (QUBO) formulations have been proposed, one using the Kronecker delta function and the other through the help of a bijective function. The interest in the use of a bijective function for a formulation was inspired by the fact that the Kronecker delta function can incur additional computational overhead and the potential to eliminate it by exploiting the inherent numerical representation of symbols. Subsequently, an evaluation based on a few simple test cases was carried out on both formulations. Here, the DWave Leap cloud solvers have been utilized and minor-embedding is implicitly handled. For evaluation purposes, a metric called the Occurrence Ratio (OR), which is based on the number of occurrences in D-Wave output was defined. Additionally, the Maximum Occurrence Ratio (MOR) was defined as the maximum value obtained for OR for any solution. With minimal hyperparameter tuning, the expected solutions were obtained for every test case, where MOR was the same as OR, except for one case (MOR ≠ OR). Since only the basic constraints have been followed to set the values for the Lagrange parameters in the Hamiltonian (QUBO), hyperparameter tuning is essential for, cases such as MOR ≠ OR. To address practical and implementation issues, an inherent decomposition strategy based on the possibility of having substrings has been elucidated to accommodate the restricted qubit count. It is deduced theoretically, that the required qubit count is mn, where m is the number of symbols in a string and n is the number of strings, whereas due to practical limitations, this number is usually higher. Conclusively, the need for further investigation on tuning the hyperparameters is emphasised.