Research Publications Authored by SLIIT Staff
Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/4195
This collection includes all SLIIT staff publications presented at external conferences and published in external journals. The materials are organized by faculty to facilitate easy retrieval.
Browse
19 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Embargo Assistant Zone – Homeschooling Assistance System based on Natural Language Processing(IEEE, 2022-12-09) Premendran, K; Bopearachchi, S.B.D.D.; Senevirathna, S.D.M.; Giridaran, S; Archchana, K; Ganegoda, D; Thelijjagoda, SAs a developing country, most people give their highest priority to education. When focusing on building an e-learning platform to improve the knowledge of students and teacher-student interactivity, the pandemic season can be mentioned as the main blocker which highly impacted the education field. Not only by considering the pandemic situation but also by addressing the concerns when it comes to teacher and student evaluation and psychological levels of students who are undergoing different difficulties, the “Home Schooling Assistance System” (Assistant Zone) has been introduced as a solution. The Assistant Zone has been initiated with three unique features which are valuable for both students and teachers. This system analyzes the strengths, weaknesses and evaluates the student performance, suggests study materials to improve themselves, provides solutions to the problems faced by the students, teachers, and parents and measures the performance of teachers based on their students, and recommends learning materials for the low-performing teachers. The Assistant Zone fulfills the targeted problems and introduces the above-mentioned three unique features with the use of Natural Language Processing (NLP) such as the BERT algorithm and Machine Learning models such as the Recurrent Neural Network, Forward Neural Network, and Gaussian Model.Publication Embargo A Mobile Gamified Application to Improve Mathematics Skills of Students from Age 9 to 11 Years(Institute of Electrical and Electronics Engineers, 2022-09-16) Karunasekara, K.K.R; Olinka, P.L.T; Kodithuwakku, K.T.D; Jayashan, B.V.P; Kahandawaarachchi, C; Attanayaka, BThe revolutionary change in technology has directly involved with e-learning systems. Mathematics is recognized as essential yet difficult for most of students. The main problem is considered as existing solutions were not provided a better solution for knowledge level enhancement of mathematics by detecting the exact knowledge levels of students. Current applications have offered permission to select students' knowledge levels by themselves or initiated from the initial level which has led to wrong identification of mathematics knowledge. As a solution for these problems gamified mobile application is provided by introducing a knowledge level detection with a whole syllabus coverage question classification and answer classification by using a Logistic Regression extra trees classifier. The knowledge improvement is performed from the exact knowledge level detected and positive emotions of students. Convolutional neural network is utilized for emotional recognition. Further mental health is improved via a Chabot as an educational encouragement and gamified application is developed based on psychology and preference around ten years by adhering to the best practices of Human-Computer Interaction. The proposed solution is considered as a multi-valued and cost-effective solution which will improve the mathematics knowledge level and good mental wellbeing of students with an appealing gamification environment.Publication Open Access User Awareness System to Diagnose Dermatological Diseases(Foundation of Computer Science, 2020-12-18) Chandrasiri, S; Weerasooriya, T; Pathivarathan, V; Thavabalasingham, N; Philipreman, K; Gunasekaran, SNowadays, humans' health is deteriorating by dermatological diseases, and the spreading rate is high. Most people are not aware of skin diseases. As they do not realize these diseases' seriousness, they try to treat with some remedies by themselves, even without knowing what the actual disease is. Nevertheless, it is not a suitable way to cure the disease, leading to future complications. So still the dermatological diseases remain as one of the main categories of common health issues. A few people prefer to use computerized systems to evaluate the disease conditions these days. Moreover, it is essential to know about the diseases to manage that condition and prevent escalation. Therefore, the proposed system is implemented to give users some knowledge about dermatological diseases as much as possible. The users can get awareness and predict skin diseases and complications from the data mining technique. The user can identify the stage of the dermatological disease by applying the classification algorithm. Furthermore, this system will also scrap web pages related to that disease from known or system verified websites. The content analysis is based on the machine learning process, especially using Neural Language Processing. Hence, the system will undeniably be useful to the users to summarize skin diseases and get concerns from a dermatologistPublication Open Access Automated Customer Care Service System for Finance Companies(NCTM, 2014-12-16) Warnapura, A. K; Rajapaksha, D. S; Ranawaka, H. P; Fernando, P. S. S. J; Kasthuriarachchi, K. T. S; Wijendra, DIn general, to obtain information about a product one should visit the company or contact the company via a phone call or some sort of a communication type, for example E-mail. Even so under normal circumstances the customer will receive the necessary information sent by a human being. There can be many disadvantages in this method. At the onset if a particular customer gives a phone call to the company the customer will have to wait for a considerable time. This is obvious because due to lack of human resources and phone lines there may be a question of customers waiting to get connected to the company line. On the other hand if a customer sends an email, the reply for the email will take time because the particular email should be perused by another human being at company in order to reply. These are few disadvantages apart from human errors that can happen. Ultimately as a result of above detrimental facts a faithful customer could get unsatisfied and lose confidence on a particular company. However, in the system that we are going to introduce, a particular customer can get any type of information in real time by the Aid of the Artificial Intelligence in the form of text/voice or E-mails. The advantages over the other method are that the customers will not have to wait for a reply, there are no space for human error and more importantly the company can use their human resources in other activities while the system takes care of the Customer care unit at least partially. Further, this system will be help to people who needs the immediate customer care assistance and will be able to get help by their own without involved human agent in another party for their assistancePublication Open Access EnglishBuddy–An approach for structured answer evaluation and feedback for O/L English language examinations in Sri Lanka.(GARI Publisher, 2017-12-31) Gurusinghe, A. M; Wijenayaka, W. K. D. P; Nawagamuwa, C. N; Priyadarshani, W. K. D. L; Gamage, M. P. A. WAt present, English has become a universal language. It may not be the most spoken language in the world, but it is the official language in a large number of countries. Proficiency in English both spoken and written has become a basic and a crucial requirement to get a decent white collar job and also to pursue higher studies or career development. Therefore, passing O/L with a good grade for English has become critical. But due to the busy life styles of students, teachers as well as parents, do not see this as a major problem and pay less attention to English compared to the other subjects. Since there is a limited time available for each subject at school, teachers might not be paying their full attention to the students who need teachers’ help. Sometimes parents also feel it is difficult to attend to the parents meetings and they might not know the actual grades of their children until the final results are given. This research provides a solution to the above problems by developing an automated system called “English Buddy” which will mark student’s structure based answers in English and help the students to learn and evaluate their knowledge alone. This web solution will be useful for teachers to upload material and check progress of the students and for students to learn and practice exercises and get feedback. It’ll be helpful for parents to be alert and follow the progress of their children. The systems is mainly build using techniques in Natural Language Processing and checked for accuracy with manual marking.Publication Open Access Speecur: Intelligent pc controller for hand disabled people using nlp and image processing(GARI Publisher, 2017-06-30) Chathumali, E. J. A. P. C; Jayasekera, J. M. U. B; Pinnawala, D. C; Samaraweera, S. M. T. K; Gamage, M. P. A. WToday computers play a major role in human lives. Even though there are very sophisticated interfaces, differently abled people find it challenging to interact with the computer. There are some applications for disabled people, but people who have disabilities in hands do not have a proper application to interact with new technologies. The aim of this project is to develop software that act as an intelligent controller to facilitate hand disabled people when interacting with a computer. Proposed intelligent solution is based on speech recognition, image processing and human computer interaction. This application is capable of moving mouse cursor with face detections, activities based on voice commands, provide user authentication by voice recognitions and give suggestions using facial emotions. The solution will be using various high end techniques in Natural Language Processing, Machine Learning and Image Processing in order to improve the computer interaction. The proposed solution will be a great solution for the hand disabled people to interact with the computers like a normal user does.Publication Open Access Virtual student advisor using NLP and automatic appointment scheduler and feedback analyser(www.ijser.org, 2016-02-02) Suvethan, N; Avenash, K; Huzaim, M. A. Q; Mathusagar, R; Gamage, M. P. A. W; Imbulpitiya, AVirtual Student Advisor is a research project that mainly concerned on addressing a comprehensive solution to overcome the difficulties faced by the academic departments of any academic institutions. According to the context, the role of the Student Advisor is focused on helping students with problems related to their academic carrier at the University and also answer the general queries made by students related to the procedures conducted at the Institute. This happens to be a tedious and a very inefficient task for the Academics as they have to repeat the same answers for many students and also students coming to meet the academic in ad hoc manner without proper appointments makes both the student and the academic face lot of problems. The Virtual Student Advisor system mainly consists of three components; A Natural language based inquiry management module, Priority wise automatic appointment scheduler and Feedback analyzer. The inquiry management module is responsible in handling user queries based on frequently asked questions. Users can get the answers for a query by entering the question in natural language. This is implemented as a mobile application as it is convenient for students to ask questions from any place at any time. The system will answer the queries and if it needs more explanation it will be directed to the relevant advisor. Auto appointment scheduler handles the student appointment requests and helps in managing the schedules. This will be accessed by both Academics to set their free time slots and the students to request an appointment according to urgency. Feedback analyzer handles the entire process of student feedback taken for each subject starting from preparing feedback forms till analyzing the collected information. The lecturer can prepare a feedback form using the question bank in the system and customize it further to fit the requirements and after conducting the feedback the system can process the data and provide reports on varies views of the data such as tables and pie charts. The system uses Natural Language Processing to handle students’ queries by tokenizing the sentences and extracting answers based on keywords and comparing synonyms WordNet lexical database. The appointment scheduler uses priority based Round Robin CPU Scheduling Algorithm to schedule the appointments according to urgency. The Virtual Student Advisor system consists of a mobile application for students and a web application for both students and lecturers to access the information and conducted their daily tasks related to academic administration effectively.Publication Open Access Evolutionary algorithm for sinhala to english translation(arXiv preprint arXiv:1907.03202, 2019-07-06) Joseph, JK; Chathurika, W. M. T; Nugaliyadde, A; Mallawarachchi, YMachine Translation (MT) is an area in natural language processing, which focus on translating from one language to another. Many approaches ranging from statistical methods to deep learning approaches are used in order to achieve MT. However, these methods either require a large number of data or a clear understanding about the language. Sinhala language has less digital text which could be used to train a deep neural network. Furthermore, Sinhala has complex rules therefore, it is harder to create statistical rules in order to apply statistical methods in MT. This research focuses on Sinhala to English translation using an Evolutionary Algorithm (EA). EA is used to identifying the correct meaning of Sinhala text and to translate it to English. The Sinhala text is passed to identify the meaning in order to get the correct meaning of the sentence. With the use of the EA the translation is carried out. The translated text is passed on to grammatically correct the sentence. This has shown to achieve accurate resultsPublication Embargo A Mobile App for Location Based Customer Notifications about Sales Offers(2019 1st International Conference on Advancements in Computing (ICAC), SLIIT, 2019-12-05) Saluwadana, R.B.; Hemachandra, K.A.N.W.; Jayasinghe, L.M.R.; Ahnaf Hassanar; Gamage, M.P.Nowadays merchants’ focus on sending specifics about their sales offers to prospective customers through electronic means. But customers are neutral about those messages if they are away from those shops. Therefore, the authors decided to implement a mobile application to send location-based sales offer notifications to customers in order to overcome this problem, with some additional features. The main features in the proposed system are to filter out sales offer details from social media, send location-based notifications containing details of offers to customers, provide personalized search predictions during search, and provide recommendations to merchants to improve their business. Modern technologies like Machine Learning (ML), Deep Learning (DL) and Natural Language Processing (NLP) are used to build the solution for this problem. The main advantage of the proposed system is that customers are attracted more towards the sales offers since they receive them when they are close by to the relevant shop. Also, merchants can reach targeted customers resulting in a more effective marketing campaign. The survey conducted proved that both customers and merchants are highly satisfied with the effectiveness of the product.Publication Open Access Simplifying Law Statements Using Natural Language Processing(SLIIT, 2016-11-16) Dharmasiri, N; Gunathilake, B; Pathirana, u; Senevirathne, S; Nugaliyadde, A; Thelijjagoda, SUnderstanding the law statements for general public is evidently complex. The research derives a computational solution on reducing the complexity of the law statements. Given a law statement, the research will use both wordnet and “LawNet” to create a simpler meaning. The research will focus on information extraction, information retrieval, question analysis and answer generation techniques to derive better meaning of law statements. The law statement will be treated as a question and the “LawNet” and wordnet will be used in as information extraction points. The law statement will be analyzed as a question; more information will be retrieved through the wordnet and “LawNet”. This process mostly acts similar to a search engine’s process. The results provide on average 80% accuracy for a 1500 dataset.