Browsing by Author "Pathirana, K.P.P.S."

Now showing 1 - 8 of 8
  • No Thumbnail Available
    Item
    Automated Essay Type Paper Marking System
    (Uva Wellassa University of Sri Lanka, 2020) Wedisa, M.A.R.; Siriwardhana, M.K.S.S.; Dayananda, P.G.C.N.; Pathirana, K.P.P.S.; Ekanayake, E.M.U.W.J.B.
    Automated paper marking is a very important research tool for the education evaluation process. Some researchers indicated that almost every study’s challenge was to get the semantic similarity of an essay rather than keyword matching. Another major problem is the lack of sufficient data that needed to train the system for a specific domain with a supervised learning approach and there are some issues with the unavailability of educator’s involvement with the scoring systems, also there were no studies that behave like a complete system. The automated scoring or evaluation for written student responses have been, and are still a highly interesting topic for natural language processing (NLP) and Machine Learning (ML) research. This study is focused on building a complete system that automates essay paper marking with a novel approach using NLP and ML. Primarily, researchers have used a hybrid approach to get the semantic similarity between two textual objects which contain word-vector-similarity, knowledge-based- similarity, and word-order-similarity. As one of the main advantages, our system uses an unsupervised learning approach, so that the system can work independently without training for a specific subject domain. The emerging of word embedding encouraged the calculation of the word-vector-similarity with Vector Space Model and cosine-similarity mechanisms. On the other hand, the word-net knowledge base was used to calculate the semantic distance between the documents and word-order-similarity played a major role in the accuracy of the final result. Also, machine learning techniques and a vast number of NLP techniques have been used for implementation. Besides, the proposed study contains an OCR to identify student's handwritten characters and also a website to easily interact with the system. In conclusion, the system was tested and evaluated with 30 samples of essays and the manual scores given by the educators. As a result, it indicated a strong positive correlation of (0.882) between manual scores and the system scores. Keywords: Automated essay scoring (AES), Natural language processing (NLP), Machine learning (ML), Optical character reader (OCR)
  • No Thumbnail Available
    Item
    Designed Artefacts for Analyzing and Evaluating Autism Spectrum Disorder (ASD)
    (Uva Wellassa University of Sri Lanka, 2020) Yapa, Y.M.U.I.; Nuska, M.N.F.; Imthath, A.M.; Pathirana, K.P.P.S.; Jayathunga, D.P.
    According to the recent statistics, 1 in 63 children are affected with Autism. Autism is a neurodevelopment disorder of early childhood, it is a condition that occurs due to the abnormal growth of mind, where these children exhibit extra-ordinary behavioral patterns. There is no well-defined treatment for Autism Spectrum Disorder (ASD), and early diagnosis is essential to manage the condition. An ICT based artifact (more specifically, a set of software) can be introduced as a novel approach, which intends to expose the child behavior. Furthermore, the outcomes of such an artifact could be used by any psychiatrist for predictions of ASD. These artifacts are designed by considering three main impaired areas of ASD which are Eye Contact, Maturity level, and Intelligence level. Therefore, the developed system is comprised of an Eye Movement Tracking tool where a common sample video is shown to the participants and a record of their eye movement is taken and this recorded data is then processed and finally displayed graphically. A module capable of identifying the Maturity Level provides a drawing canvas where participants are allowed to draw shapes and the analysis is done by the way they draw correct shapes with time in graphs. Moreover, an Intelligence Level Measuring Tool compromised with color and number-based activities is used and their responses are taken for decision making. Besides, these artifacts are capable of giving an analysis by comparing both ASD patients and a Neurotypical person. Testing and evaluation of the system were done with three (3) ASD patients and ten (10) Neurotypical persons from the age groups of 3-5 years. This experiment showed that, computer-based software tools are effective for acting as a platform to provide data and for taking decisions in ASD predictions. Keywords: Autism Spectrum Disorder (ASD), Eye contact, Intelligence level, Maturity level, Neurotypical person
  • No Thumbnail Available
    Item
    Enhancement of English Language Speech and Comprehension Through Means of Virtual Reality for Sri Lankan Context
    (Uva Wellassa University of Sri Lanka, 2020) Jayarathna, H.M.P.P.; Priyanga, E.A.I.; Jayasinghe, K.C.; Senanayake, S.H.D.; Wimaladarma, S.T.C.I.; Pathirana, K.P.P.S.
    Speaking is one of the essential skills needed to be developed by any English learner. But the English curriculum taught by Sri Lankan schools often focuses on providing English vocabulary, grammar, and comprehension but pay less attention to enhance oral communication skills. This leads to a lack of practice in spoken English. Therefore, that affects a wide spectrum of applications such as business communication, presentation of creative work, etc. Other problems are anxiety, fear, lack of confidence, and nervousness of speaking English. This research focuses on designing a Virtual Reality (VR) based application for the above-mentioned problems and it facilitates a more convenient and yet natural experience through the use of techniques such as VR, Natural Language Processing (NLP), etc. to expand the English oral and understanding ability. This VR based solution will enhance the oral practice of the language and reduce the lack of confidence while improving the speech and comprehensive skills of Sri Lankan school students. Users can practice the application anywhere with common real-world scenarios such as conversing during a doctor’s appointment, buying dresses from a shop, etc. This game-based learning tool helps to evaluate the users and they can get scores according to their performance. At the same time, it facilitates to expand their English speech and comprehension ability. To measure the effectiveness of this application, user evaluation was done as a pre-experimental method with a one-shot case study with the use of pre-test and post-test design. A total of 30 local students following the local English language curriculum between grades eight and ten were selected as a sample for this purpose. The t-test analysis showed a value of 2.34 alongside a table value of 2.131 which depicts a clear correlation between the usage of the application. Significant enhancement of oral and comprehensive skills of the users was observed through the evaluation. Keywords: Virtual reality, Natural language processing, Voice recognition, Speech synthesis, Artificial intelligence
  • No Thumbnail Available
    Item
    High Tech Vision to Detect Currency Denomination and Virtual Wallet to Retrieve the Monetary Position for Visually Debilitated People
    (Uva Wellassa University of Sri Lanka, 2020) Perera, N.B.L.K.; Franciscus, A.U.; Wickramasinghe, M.G.R.D.; Jayasekara, N.E.C.; Pathirana, K.P.P.S.; Jayakody, J.A.V.M.K.
    The transformation of currency notes and coins denomination recognition to an automated system as a solution for visually debilitated individuals to overcome the difficulties facing when handling monetary transactions. This research presents a model to detect currency notes and coins to visually debilitated individuals and to retrieve the current monetary position of them as per their obligation and provide audio output in the Sinhala language. The general procedure of the system includes digital image processing, convolutional neural network, voice identification algorithm, and monetary position calculation algorithm. Sri Lanka currency notes and coins images were captured in a wide variety of environments, in association with lighting conditions and background to make the data set, using the image preprocessing technique. The YOLOv2, R-CNN network model which is a high speed, real-time object detection algorithm to verify objects as currency notes and coins. Then by using Keras Xception model, predict images, feature extraction and fine-tuning have been done to train the data set. The Computer vision used to improve machine perception to retrieve real-time detection. The detected currency notes or coins denomination is provided as an audio output, then retrieves the obligation of the user, which is whether to debit, credit or to retrieve the current monetary position. The monetary position provides audio output in the virtual wallet as a substitute for a realworld wallet since impairments have a scarcity in memorizing their actual balance. The study revealed a system to detect and retrieve the currency denomination and monetary position of blind individuals with the overall accuracy rate of 100% in algorithm experiments. Keywords: Visually debilitated individuals, Currency recognition, Virtual wallet, monetary position
  • No Thumbnail Available
    Item
    An Image Processing Application for Diagnosing Acute Lymphoblastic Leukaemia (ALL)
    (Uva Wellassa University of Sri Lanka, 2021) Rathnayake, R.M.S.K.K.; Piyumali, R.W.S.U.; Withanage, W.S.U.; Pathirana, K.P.P.S.; Wilson, R.S.I.
    Acute Lymphoblastic Leukaemia is a fatal disease that affects white blood cells and bone marrow in the human body. Every year, considerably a large number of adolescents and children become victims of this type of leukaemia. The early detection of this disease directly affects the recovery rate of the patients. In the manual process, pathologists can identify Acute Lymphoblastic Leukaemia and the accuracy of the prediction may rely upon their experience. Hence this research has proposed an image processing approach for early detection of Acute Lymphoblastic Leukaemia cells to prevent the spreading of cancer, enabling the medical experts to initiate the treatment without any delay and increase the recovery rate of such patients. For that, microscopic blood sample images were analyzed considering the features such as color, shape, presence of nucleoli, and nucleon to a cytoplasmic ratio of the cells separately using three Conventional Neural Networks (CNNs). Based on that, the Acute Lymphoblastic Leukaemia cells were identified and classified as either Acute Lymphoblastic Leukaemia or healthy. Compared to the laboratory testing methods, this approach obviously leads to early detection of Acute Lymphoblastic Leukaemia with an accuracy of 94.57% that has been confirmed by the domain experts. The proposed approach is an effective and less expensive method that would assist doctors to get fast and accurate results. Hence the originality of this research was to identify the presence of Acute Lymphoblastic Leukaemia cells in the microscopic blood sample images and classify them as either Acute Lymphoblastic Leukaemia or healthy by identifying the features of the Acute Lymphoblastic Leukaemia cells separately. Moreover, this research has found that Conventional Neural Networks (CNN) is the most suitable Neural Network to identify Acute Lymphoblastic Leukaemia using image processing technique. Keywords: Acute Lymphoblastic Leukaemia; white blood cells; conventional neural networks; Image Processing; Machine Learning
  • Thumbnail Image
    Item
    Medical Image Analysis with an Innovative Intelligent Computer Program
    (Uva Wellassa University of Sri Lanka, 2010) Pathirana, K.P.P.S.
    The pre-defined major research objective was uniquely identifying an image captured from a medical instrument such as electronic microscope (microscopic view). It was proposed that, the technique optical patterns recognition, one of the key applications of artificial neural networks (ANN) will be used. Furthermore, also mentioned that the researcher will study the feasibility of using genetic algorithms (GA) to enhance the efficiency of ANN. In brief, the research is developing a computer program (software) with its ultimate objective, analyze a digital image and recognize it. (Optionally —Interpret it.) Many experiments were carried-out to achieve these objectives, started with implementing an algorithm (a neural network) to recognize optical patterns. Succeeded with the selected software to implement and simulate basic algorithms -`Wolfram Mathematica Version 6.0'. The prototype-level NN architecture comprised of 3 layers, successfully trained with `Back-propagation algorithm' and was able to recognize some pre-trained images of Sinhala characters (for algorithms testing purposes only). This program was then interfaced with a GUI created from VB.net and worked fine. (A Sinhala OCR-Optical Character Recognizer — a by-product) In the next step, the training algorithm was optimized with an innovated Genetic Algorithm (GA), tested and obtained reports verifying that the GA works as expected and it significantly reduces the errors of outputs. Then, after an in-depth study, the type of medical image (from which instrument they are being captured from) was selected. That was color images from TEM (Transmission Electron Microscopy) (http://en.wikipedia.org/wiki/Transmission_electron_microscopy). The images analyzed by the program are - more specifically, human blood viruses those can be recognized by its specific optical patterns.
  • No Thumbnail Available
    Item
    Medical Image Analysis with an Innovative Intelligent Computer Program
    (Uva Wellassa University of Sri Lanka, 2010) Pathirana, K.P.P.S.; Dissanayake, B.A.K.
    The objective of the research is computer aided desicion making by analyzing color images captured from a medical instrument called TEM (Transmission Electron Micro-scope) with an Artificial Intelligence (AI) enabled computer program. When the images are analyzed specifically, viruses can be recognized by their optical patterns. The final outcome of the research is an intelligent computer program with Graphical User Interfaces (GUI) to analyze Transmission Electron Micrographs (TEM), to recognize the viruses and generate automated reports (Interpretation). The technique optical pattern recognition and classification, one of the major applications of Artificial Neural Networks (ANN or NN) was applied as the key algorithm of the computer program. Since NN is a mathematical model of complicated human brain, the application derived from NN is integrated with Al and produces intelligent decisions regarding the specified trained task, TEM analysis. This Al enabled system has many significant advantages compared to traditional non-Al image analysis software, such as recognizing mutated forms of viruses, ability to analyze even distorted images, dynamically expanding knowledgebase (self-improving knowledge)etc. In other words, the final outcome of this research is a Virtual Virologist —specialist in human blood viruses, known as an expert system in Al field. Furthermore, the feasibility of using Genetic Algorithms (GA) to enhance the efficiency of NN was also exprimented. The training algorithm of NN was optimized with an innovative GA, tested and obtained reports verifying that the GA behaves even better than expected and reduces the output error. The system developed was tested with large number of relevant images and has produced expected results. Key words: Genetic algorithm, Artificial neural networks, Transmission electron micro-graph, Artificial intelligence
  • No Thumbnail Available
    Item
    Open BevyBot 2020 – An Open Source Low Cost Educational Robot for Effective Learning
    (Uva Wellassa University of Sri Lanka, 2020) Silva, B.P.D.; Karunarathne, K.R R.; Shaya, K.; Pathirana, K.P.P.S.; Senanayake, S.H.D.
    The modern world with lack of availability of an effective learning platform for students who are tired of traditional learning techniques. In the present day, basic mathematics is the most valuable fact for primary level students to improve their problem-solving skills. This study is carried out to develop a cost-effective open-source robot with a rewarding system for the children of the age group between five and ten years. The open-source robot chassis designed with computer-aided designing which can print by using 3D printing technology. In this research, a Java-based library is developed to communicate with the microcontroller of the robot and the hardware of the smartphones. And also, there is another problem that they have to spend more strive to build a robot as it is not a work that can be done in less time with a low budget. Throughout this study, present solutions such as Cosmo, Poppy, Q.bo-one are considered to gather information. But these robots are costly and also lack of opportunities to use for education purposes and that robots need external hardware sensors. But in the modern world, most people own a smartphone and it contains the above hardware (Processor unit, Sensors.). If can reuse that mobile phone hardware as robot hardware it will much cost-effective. Researchers decided to combine three ideas of open-source hardware/software, 3D printing technology, and reuse mobile phone hardware. As the final solution researchers build an open-source robotic platform with reusing mobile phone hardware combining the android library and finally they build an educational robot to evaluate the platform. Furthermore, researchers analyse the effectiveness of the built educational robot by providing a questionnaire to the students (age group 5-10 years) & getting feedback from them. Researchers plan to use PCB designs as hardware circuits into one platform which reduces the circuit space and commercialization of the product in the future. Keywords: Educational robots, Open source, 3D printing technology, Low cost, Mobile phone