Browsing by Author "Sumathipala, S"

Now showing 1 - 8 of 8

item: Conference-Full-text
Augmented reality based breadboard circuit building guide application
(2018) Thiwanka, N; Chamodika, U; Priyankara, L; Sumathipala, S; Weerasuriya, GT; Wijesiriwardana, CP
Building circuits on breadboards is an activity which requires a lot of attention and thinking. If there is a way to guide this process by using modern technologies, the learning process can be made more effective and interactive. This study proposes a solution that provides students with an augmented reality visualization of the expected circuit on a breadboard before they actually make the circuit. The proposed system can be divided into four main modules based on their functionality (a) extracting possible information from the electronic components, (b) scanning circuit diagrams for identifying circuit symbols and their connectivity, (c) finding the appropriate arrangement of the electronic components on the breadboard and (d) using augmented reality to visualize the circuit on a breadboard. This solution provides an innovative approach to facilitate the learning process of students by making electronic circuit building interesting and interactive.
item: Conference-Full-text
Explainable ai techniques for deep convolutional neural network based plant disease identification
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa., 2023-12-07) Kiriella, S; Fernando, S; Sumathipala, S; Udayakumara, EPN; Piyatilake, ITS; Thalagala, PD; Ganegoda, GU; Thanuja, ALARR; Dharmarathna, P
Deep learning-based computer vision has shown improved performance in image classification tasks. Due to the complexities of these models, they have been referred as opaque models. As a result, users need justifications for predictions to enhance trust. Thus, Explainable Artificial Intelligence (XAI) provides various techniques to explain predictions. Explanations play a vital role in practical application, to apply the exact treatment for a plant disease. However, application of XAI techniques in plant disease identification is not popular. This paper discusses the key concerns and taxonomies available in XAI and summarizes the recent developments. Also, it develops a tomato disease classification model and uses different XAI techniques to validate model predictions. It includes a comparative analysis of XAI techniques and discusses the limitations and usefulness of the techniques in plant disease symptom localization.
item: Conference-Full-text
Generating entity relationship diagram from requirement specification based on nlp
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Kashmira, PGTH; Sumathipala, S; Wijesiriwardana, CP
An entity relationship data model is a high level conceptual model that describes information as entities, attributes relationships and constraints. Entity relationship diagrams to design the database of the software. It involves a sequence of tasks including extracting the requirements, identifying the entities, their attributes, the relationship between the entities, constraints and finally drawing the diagram. As such entity relationship diagram design has become a tedious task for novice designer. This research addresses the above issue, proposes a Natural Language Processing based tool which accepts requirement specification written in English language and generates entity relationship diagram.
item: Conference-Full-text
A study on autonomous entering into narrow path using a mobile robot
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Matsumoto, Y; Premachandra, C; Sudantha, BH; Sumathipala, S; Wijesiriwardana, CP
This research aimed to find a narrow path identification method using autonomous movement of various robots including wheel robots. It was used to detect obstacles by sensors, obstacle avoidance behavior and entering behaviors to the narrow path using a compact wheel robot ZUMO. Fuzzy logic methods were used in order to acquire accurate angle information at the time of entrance into a narrow path. Furthermore, both simulation and actual verification experiments were conducted to confirm the effectiveness of the system.
item: Conference-Full-text
A study on mobile robot control by hand gesture detection
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Ikegami, S; Premachandra, C; Sudantha, BH; Sumathipala, S; Wijesiriwardana, CP
Recently, method of controlling machine with natural control called Natural User Interface has attracted attention. particularly, human hand is considered to be communication media that can communicate information most naturally. In this paper, user hand is detected by following acquired skin color of the user's face, through face detection. After detecting the user hand, hand gesture is determined. Finally, a robot is controlled to have multi movements according to detected several hand gesture patterns. We succeeded in acquiring the skin color information from user's face and detecting both user’s hand and hand gesture following that. In this paper, we developed a mobile robot to conduct experiments. We also confirmed that robot can be controlled by human hand gesture.
item: Conference-Full-text
System for detecting student attention pertaining and alerting
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Vettivel, N; Jeyaratnam, N; Ravindran, V; Sumathipala, S; Amarakecrthi, S; Wijesiriwardana, CP
Concentration is the ability to focus the mind on a specific context at a time. This is important to do any activities efficiently such as learning and driving. Especially, for students, it is tough to be stay focused because of the distractions. A single moment of drifting in mind may cause a significant impact on student’s performance. Therefore finding that moment and alert the student to regain attention, would help him to improve the ability to be concentrated while learning. Several types of researches have been conducted to find out the connection between concentration and human related parameters such as heart rate variability, brain waves, and facial expressions while learning. We propose a methodology to combine these three parameters, expected to overcome the limitations of one parameter by another. The extracted features from each collected data from the relevant sensors are fed into the classification models. As per the initial experiments, the primary relationships were derived with separate machine learning models for each parameter.
item: Conference-Full-text
Text summarization for tamil online sports news using nlp
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018-12) Priyadharshan, T; Sumathipala, S; Wijesiriwardana, CP
Text summarization plays an important problem in natural language understanding and information retrieval. Automatic text summarization get much more attention by people presently because it is efficiently and effectively serve time in decision making process even for day to day life. Presently deep learning models get more attention than the traditional approaches. The primary objective of this research work is to propose a methodology to address the problem of summarization for Tamil sports news which can automatically create extractive summary for the news data with the use of Natural Language Processing (NLP) and a generic stochastic artificial neural network. Features such as sentence position, sentence position related to paragraph, number of named entities, term frequency and inverse document frequency and Number of numerals are employed to construct the feature matrix for each sentence and Restricted Boltzmann Machine is used to improve those features while enhancing the accuracy without loosing the main idea of the text. Experimentation is carried out using Online Tamil sports news and ROUGE tool kit is used to evaluate the recall, precision and F-measure for the summary generated by both the human experts and the system.
item: Conference-Full-text
Word level language identification of code mixing text in social media using nlp
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Shanmugalingam, K; Sumathipala, S; Premachandra, C; Wijesiriwardana, CP
Understanding social media contents has been a primary research topic since the dawn of social networking. Especially, contextual understanding of the noisy text, which is characterized by a high percentage of spelling mistakes with creative spelling, phonetic typing, wordplay, abbreviations, and Meta tags. Thus, the data processing demands a more complex system than traditional natural language processors. Also people easily mixing two or more languages together to express their thoughts in social media context. So automatic language identification at word level become as necessary part for analyzing the noisy content in social media. It would help with the automated analysis of content generated on social media. This study uses Tamil-English code-mixed data from popular social media posts and comments and provided word level language tags using Natural Language Processing (NLP) and modern Machine Learning (ML) technologies. The methodology used for this system is a novel approach implemented as machine learning classifier based on features such as Tamil Unicode characters in Roman scripts, dictionaries, double consonant, and term frequency. Different machine learning classifiers such as Naive Bayes, Logistic Regression, Support Vector Machines (SVM), Decision Trees and Random Forest used in training and testing. Among that the highest accuracy of 89.46% was obtained in SVM classifier.