Research Publications
Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/4194
This main community comprises five sub-communities, each representing the academic contribution made by SLIIT-affiliated personnel.
Browse
60 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Open Access Machine Learning Modelling of the Relationship between Weather and Paddy Yield in Sri Lanka(Hindawi, 2021-05) Ekanayake, P; Rankothge, W; Weliwatta, R; Jayasinghe, J. M. J. WThis paper presents the development of crop-weather models for the paddy yield in Sri Lanka based on nine weather indices, namely, rainfall, relative humidity (minimum and maximum), temperature (minimum and maximum), wind speed (morning and evening), evaporation, and sunshine hours. The statistics of seven geographical regions, which contribute to about two-thirds of the country’s total paddy production, were used for this study. The significance of the weather indices on the paddy yield was explored by employing Random Forest (RF) and the variable importance of each of them was determined. Pearson’s correlation and Spearman’s correlation were used to identify the behavior of correlation in a positive or negative direction. Further, the pairwise correlation among the weather indices was examined. The results indicate that the minimum relative humidity and the maximum temperature during the paddy cultivation period are the most influential weather indices. Moreover, RF was used to develop a paddy yield prediction model and four more techniques, namely, Power Regression (PR), Multiple Linear Regression (MLR) with stepwise selection, forward (step-up) selection, and backward (step-down) elimination, were used to benchmark the performance of the machine learning technique. Their performances were compared in terms of the Root Mean Squared Error (RMSE), Correlation Coefficient (R), Mean Absolute Error (MAE), and the Mean Absolute Percentage Error (MAPE). As per the results, RF is a reliable and accurate model for the prediction of paddy yield in Sri Lanka, demonstrating a very high R of 0.99 and the least MAPE of 1.4%.Publication Embargo User-friendly Enhanced Machine Learning-based Railway Management System for Sri Lanka(IEEE, 2021-12-06) Mihiranga, G.L.V; Weerasooriya, W. K. M; Palliyaguruge, T. L. P; Gunasekera, P. N. G; Gamage, M. P; Kumari, S. P. KThe railway service is a convenient and low-cost transport method in Sri Lanka, widely employed by both local and foreign passengers. Major railway lines in Sri Lanka cover unique and very different areas in the country. For example, the Northern province's weather and geography conditions significantly differ from Southern or Central provinces. Majority of the tourists lack understanding in identifying appropriate or attractive places that best suits them, close by to the Railway Stations. Therefore, a passenger needs to spend more time identifying their railway tour destinations. When passengers are booking tickets, even though they are able to reserve seats beforehand, they are unable to reserve a specific seat. Also, there is no process to identify the most suitable seat for them amidst many other travelers, especially if they are travelling alone. Considering the aforementioned, authors propose a more innovative and user-friendly system for the Railway Department of Sri Lanka. Depending on various passenger attributes the system is capable of suggesting a travel plan with railway lines which cover most suitable destination suggestions; identifying the best seats with a relaxing atmosphere; providing an interactive chatbot to satisfy user queries on specific location information; and facility for 24×7 user interaction. A travel plan can save passengers time and allows them to identify the desired railway line and relevant attractions without much hassle. And they are saved of an unpleasant experience through the suggestion of the best seating location. Machine Learning and Deep Learning technologies are used in developing the proposed system.Publication Open Access Tievs: Classified Advertising Enhanced Using Machine Learning Techniques(IEEE, 2021-12-06) Ranawake, D; Bandaranayake, S; Jayasekara, R; Madhushani, I; Gamage, G; Kumari, SThe scarce use of tangible periodicals led to a consistently soaring popularity of online classified advertising. Nevertheless, existing platforms retain complications. Most recommendation systems are built with conventional technologies that are less scalable, less accurate, and having high latency processes. Moreover, customers find it tiring when clarifying a reliable, precise price value for items they are trying to sell through the classified advertising system. Additionally, strict validation techniques to identify and prevent fraudulent content or images from being published in the advertising portals have been neglected. Therefore, authors have inaugurated a superior classified advertising system, Tievs, as a solution, by appraising said predicaments. It wields a flexible, process-simplifying, concurrency-induced recommendation breakthrough implemented from Universal Sentence Encoding incurred Natural Language Processing and Deep Learning routines. Furthermore, an innovative price prediction system having a supervised regression-based ensemble model forged ensuing a comparative study, having excellent accuracy in proactively predicting item prices as to cater hassles faced by customers, was satisfied. Light Gradient Boosting classifier-driven fake description analysis and a Convolution Neural Network powered figure deception recognition system were introduced, which gained prodigious precision with moral clarity in fraud detection and prevention. Hence, the proposed solution's objective of surpassing former classified advertising systems in delivering customers' necessities, using the most lucrative, time-saving, human-centric, and error-preventive approaches, was accomplished. It was affirmative by the positively responded questionnaire regulated among prospective users by the authors.Publication Embargo Machine Learning-Based Skin And Heart Disease Diagnose Mobile App(IEEE, 2021-07-01) Tharushika, G. K. A. A; Rasanga, D. M.T; Weerathunge, I; Bandara, PThis research aims to develop a Mobile app for predicting major diseases we have to face nowadays. These days the heart disease is the main source of death around the world. It is a complex task to predict a heart attack with a doctor because more experience and knowledge are needed. Sometimes it may be gastritis or asthma symptoms. Also, the following most common disease is a skin disease. Most people have some skin disease, and they don’t even have time to check it from a medical centre. These diseases led to deadly cancers kind of things. Implementing the Smart health care application, the skin disease classification and treatment, and the heart disease predictions can be made domestically. The application is taken images of skin disease through the device camera. It classifies the disease with the Keras ResNet trained to classify the accuracy as eighty-seven point eighty-three as a percentage. The heart disease prediction module takes 14 different attributes that can access by the personal and predict the heart disease probability with the model of sklearn KNeighborsClassifier is trained as a percentage with an accuracy of eighty-three point nine. The application was developed on top of the android platform with the SQL Lite database integration.Publication Embargo Machine Learning-Based Skin And Heart Disease Diagnose Mobile App(IEEE, 2021-07-01) Tharushika, G. K. A . A; Rasanga, D. M. T; Weerathunge, I; Bandara, PThis research aims to develop a Mobile app for predicting major diseases we have to face nowadays. These days the heart disease is the main source of death around the world. It is a complex task to predict a heart attack with a doctor because more experience and knowledge are needed. Sometimes it may be gastritis or asthma symptoms. Also, the following most common disease is a skin disease. Most people have some skin disease, and they don’t even have time to check it from a medical centre. These diseases led to deadly cancers kind of things. Implementing the Smart health care application, the skin disease classification and treatment, and the heart disease predictions can be made domestically. The application is taken images of skin disease through the device camera. It classifies the disease with the Keras ResNet trained to classify the accuracy as eighty-seven point eighty-three as a percentage. The heart disease prediction module takes 14 different attributes that can access by the personal and predict the heart disease probability with the model of sklearn KNeighborsClassifier is trained as a percentage with an accuracy of eighty-three point nine. The application was developed on top of the android platform with the SQL Lite database integration.Publication Embargo EasyTalk: A Translator for Sri Lankan Sign Language using Machine Learning and Artificial Intelligence(IEEE, 2020-12-10) Manoj Kumar, D; Bavanraj, k; Thavananthan, S; Bastiansz, G. M. A. S; Harshanath, S. M. B; Alosious, JSign language is used by the hearing-impaired and inarticulate community to communicate with each other. But not all Sri Lankans are aware of the sign language or verbal languages and a translation is required. The Sri Lankan Sign Language is tightly bound to the hearing-impaired and inarticulate. The paper presents EasyTalk, a sign language translator which can translate Sri Lankan Sign Language into text and audio formats as well as translate verbal language into Sri Lankan Sign Language which would benefit them to express their ideas. This is handled in four separate components. The first component, Hand Gesture Detector captures hand signs using pre-trained models. Image Classifier component classifies and translates the detected hand signs. The Text and Voice Generator component produces a text or an audio formatted output for identified hand signs. Finally, Text to Sign Converter works on converting an entered English text back into the sign language based animated images. By using these techniques, EasyTalk can detect, translate and produce relevant outputs with superior accuracy. This can result in effective and efficient communication between the community with differently-abled people and the community with normal people.Publication Open Access Development of Cyber Threat Intelligence System in a SOC Environment for Real Time Environment(Department of Computing and Information Systems, Faculty of Applied Sciences, Sabaragamuwa University of Sri Lanka, 2021-02-24) Varatharaj, A; Rupasinghe, P. L; Liyanapathirana, CNow a days, Information Communication Technology (ICT) plays an important role in the world. In IT, Cyber Security holds a vast place. Cyber Threat Intelligence (CTI) leads the significant place within Cyber Security, as many Cyber Threats need to be faced every day by a particular organization. Security Operation Center (SOC) helps to monitor and analyze an organization’s security position in Real Time. This paper proposes about the Cyber Threat Intelligence framework in a SOC Environment in Real Time. The proposed framework contains of three layers, which are built above Security Onion. The Layer 1 comprises of input data from online and offline sources. In Layer 2, implemented two components namely Filter data and Cut down data, which receive the data from Layer 1. Finally, in Layer 3 delivers a detailed report. As the input for the Layer 1, Financial Datasets is used. These Financial Datasets, which helps in order to detect the Financial Frauds. Machine Learning is used to train the model. By implementing CTI System in an organization, it helps to gain predictive output regarding the upcoming threats. Also, it helps to ensure the reputation of an organization by establishing trust between the users. Helps to increase the number of customers to an organization. The above are the advantages gained by a particular organization by having a CTI System.Publication Open Access A Review of Data Mining Methods for Educational Decision Support(Faculty of Graduate Studies, University of Kelaniya, Sri Lanka, 2016) Kasthuriarachchi, K. T. S; Liyanage, S. RData mining is a computer based information system which is devoted to scanning huge data repositories, generate information and discover knowledge. It attempts to uncover data patterns, organize information of hidden relationships, structure association rules and many more operations that cannot be performed using traditional computer based information systems. Therefore, data mining outcomes represent a valuable support for decisions making in various industries and education is one domain that can benefit from data mining. Application of data mining in education is living in its spring time and preparing for a hot summer season. Educational data mining emerges as a paradigm oriented to design models, tasks, methods, and algorithms for exploring data from educational settings. Educational Data Mining develops and adopts statistical methods, machine- learning and data mining methods to study educational data generated basically by students and educational instructors. The main goal of applying data mining in education is largely to improve learning by enabling data driven decision making for improve current educational practices and learning materials. Educational knowledge discovery, in data mining point of view can be seen as a similar process of applying the general knowledge discovery and data mining process and in experimental point of view, it can be seen as an iterative cycle of hypothesis formation, testing and refinement which not just turn data into knowledge but, also to filter the mined knowledge for decision making. There are many applications in education arena that have been resolved using data mining. There are more research studies have also been conducted under various educational problem categories and also there are a number of frequently used data mining methods use in Educational Data Mining. Various open source and commercial tools are available to apply data mining methods on the educational data. This study focuses on the identification of various educational problem domains where data mining methods can be applied and to study the suitability of the available data mining methods and the tools to perform Educational Data Mining in Sri Lankan Educational Institutes. The knowledge discovered by this review is expected to generate meaningful insight and provide guidance for important decisions made by educatorsPublication Open Access Recommendations for Students in Higher Education: A Machine Learning Approach.(International Postgraduate Research Conference (IPRC) , Faculty of Graduate Studies, University of Kelaniya, 2017) Kasthuriarachchi, K. S. T; Liyanage, S. REducational Data Mining is a rising discipline in Data Mining setting which concentrated on creating systems for investigating one of a kind data that starts from educational settings, and utilizing those procedures to better comprehend students and the settings which they learn in. There were numerous potential circumstances for applying data mining in education, such as; predicting the performance of students in education domain, advancement of student models, making methodologies for instructive help, settling on decisions to growing better learning systems, upgrading the execution of students and lessening the dropout rate of students and so on. There were sure examinations directed in dissecting students' data to foresee the execution in light of data mining approaches utilizing machine learning algorithms. However, a few of them were guiding the students using the recommendations of educators to success in their academic lives. The key objective of this research is to provide educators‘ recommendations to students in higher education through data analysis using machine learning algorithms. In this experiment, the data about more than 3000 students with eight attributes; age, gender, A/L Stream, A/L English Grade, does the student has repeat modules, GPA of Semester1, GPA of Semester 2 and Pass status of year 1 were included into the research sample who registered and were following their first academic year of an Information Technology degree in an institute. Three classification type machine learning algorithms were used to build the predictive model. They were Naïve Bayes algorithm, Decision Tree algorithm and Support Vector Machine algorithm. The accuracy of the models built by each algorithm have been tested against each other to identify the best model and extracted the most influencing/ important attributes in the model to predict the final grade (pass/ fail) in the end of first year of the students. Accordingly, the accuracy measures of Naïve Bayes, Decision tree andSupport Vector Machine were recorded as 74.67%, 74.01% and 74.01% respectively and it was clear that all three algorithms were holding almost same accuracy level. However, the model generated by Naïve Bayes algorithm has been selected since it was outperformed the rest. Then rank features by importance method was used as the feature selection method to identify the most influencing factors of the predictive model. As the result of it, past repeat modules, GPA of Semester1, GPA of Semester 2 were extracted as the most influencing attributes. Furthermore, these attributes were tested using correlation analysis to measure the significance of the relationship with the target attribute. According to this study, the educators will be able to recommend the students to score good marks for assessments of the subjects to obtain a better GPA to semester 1 and semester 2 without failing the modules to successfully complete the first year of the degree course which make more beneficial for educators as well as students to be success.Publication Embargo Performance evaluation on machine learning classification techniques for disease classification and forecasting through data analytics for chronic kidney disease (CKD)(IEEE, 2017-10-23) Gunarathne, W. H. S. D; Perera, K. D. M; Kahandawaarachchi, K. A. D. C. PChronic Kidney Disease (CKD) is considered as kidney damage which lasts longer than three months. In Sri Lanka, CKD has become a severe problem in the present days due to CKD of unknown aetiology (CKDu) that can be seen popularly in North Central Province. Identifying CKD in the initial stage is important to provide necessary treatments to prevent or cure the disease. In this work main focus is on predicting the patient's status of CKD or non CKD. To predict the value in machine learning classification algorithms have been used. Classification models have been built with different classification algorithms will predict the CKD and non CKD status of the patient. These models have applied on recently collected CKD dataset downloaded from the UCI repository with 400 data records and 25 attributes. Results of different models are compared. From the comparison it has been observed that the model with Multiclass Decision forest algorithm performed best with an accuracy of 99.1% for the reduced dataset with the 14 attributes.