Browsing by Author "Pemadasa, N"

Now showing 1 - 8 of 8

Embargo
Analyzing Payment Behaviors And Introducing An Optimal Credit Limit
(IEEE, 2019-12-05) Bandara, H. M. M. T; Samarasinghe, D. P; Manchanayake, S. M. A. M; Perera, L. P. J; Kumaradasa, K. C; Pemadasa, N; Samarasinghe, P
Identifying an optimal credit limit plays a vital role in telecommunication industry as the credit limit given to customers is influence on the market, revenue stabilization and customer retention. Most of the time service providers offer a fixed credit limit for customers which may cause customer dissatisfaction and loss of potential revenue. Therefore, it is essential to determine an optimal credit limit that maintains customer satisfaction while stabilizing the company revenue. Clustering algorithms were used to group customers with similar payment and usage behaviors. Then the optimal credit limit derived for each cluster is applicable to all the customers within the cluster. In order to identify the most suitable clustering algorithm, cluster validation statistics namely, Silhouette and Dunn indexes were used in this research. Based on the scores generated from these statistics KMeans algorithm was chosen. Furthermore, the quality of the KMeans clustering was evaluated using Silhouette score and the Elbow method. The optimal number of clusters are identified by those validation statistics. The significance of this approach is that the optimal credit limits generated by these clustering models suit dynamic behaviors of the customer which in turn increases customer satisfaction while contributing to reducing customer churn and potential loss of revenue.
Embargo
Automated Diabetic Retinopathy Screening With Montage Fundus Images
(IEEE, 2020-12-10) Kumari, S; Padmakumara, N; Palangoda, W; Balagalla, C; Samarasingha, P; Fernando, A; Pemadasa, N
Diabetic retinopathy (DR), also known as diabetic eye disease is one of the major causes of blindness in the active population. The longer a person has diabetes, higher the chances of developing DR. This research paper is an attempt towards finding an automatic way to staging DR using montage eye images through artificial intelligence (AI). Convolutional neural networks (CNNs) play a big role in DR detection. Using transfer learning and hyper-parameter tuning DR staging is analyzed through different models. VGG16 gave the highest classification accuracies for the stages Proliferative DR (PDR) & Non-proliferative DR (NPDR). The highest level of NPDR is severe DR which achieved 94.9% classification accuracy (CA) & special features like cotton wool & laser treatment performed at 83.3% CA for each. Moreover, by using patient's history data such as age, right eye & left eye value accuracies & diabetic diagnosed year, system can predict the DR stages. That predictive model has achieved the best CA of 94 % by using Xgboost classifier. Overall, a fully functional app has been developed to detect DR stages with Montage Fundus images using AI.
Embargo
Deep learning based flood prediction and relief optimization
(IEEE, 2019-12-05) Pathirana, D; Chandrasiri, L; Jayasekara, D; Dilmi, V; Samarasinghe, P; Pemadasa, N
Flood is a major natural disaster that occurs recurrently in Sri Lanka. It is important to stay on alert and get early preparations to avoid unnecessary risks that cause damage to both life and property. This project developed a flood assistance application “DHARA” to support early flood preparation and flood recovery process. DHARA mobile application facilitates river water level prediction, safest evacuation route suggestion and provides relevant warnings and alert notifications and the web application provides affected area detection, victim and relief estimation to assist flood recovery management. This system is developed as a mobile application and a web application. A recurrent neural network architecture named Long Short Term Memory (LSTM), Convolutional Neural Network (CNN), a path finding algorithm namely A star (A*) algorithm and a clustering technique named Fuzzy Clustering are used for the development of the system. The system is verified with sample data related to “Wellampitiya” and “Kaduwela” area based on river “Kelanl”. The river water level prediction model has successfully predicted the water level 4 hours in advance. The verification results of the river water level prediction showed a satisfactory agreement between the predicted and real records with 85.4% accuracy.
Open Access
Machine learning approach for predicting career suitability, career progression and attrition of IT graduates
(IEEE, 2021-12-02) Bannaka, B. M. D. E; Dhanasekara, D. M. H. S. G; Sheena, M. K; Karunasena, A; Pemadasa, N
The IT industry in Sri Lanka is associated with a massive work force consisting of skillful professionals and it also provides many job opportunities for fresh graduates at the present. For a fresh graduate entering the IT industry there is a wide variety of job opportunities available and in order to have a satisfactory and rewarding career they should identify the most suitable career for them. On the other hand, employees change their careers and regularly seeking for career advancements and more benefits while the employers struggle to retain employees. Under such circumstances, this research focuses on developing a career mentoring system which comprises of the prediction of career suitability, career and salary progression, and employee attrition to assist IT employees to achieve career goals by overcoming barriers in their career path. For this purpose, data are collected from IT employees, and several models were implemented using classification algorithms such as XGBoost, Random Forest, Support Vector Machine, K-Nearest Neighbors, Decision tree, Naive Bayes, and their performance are compared using accuracy, precision, recall, and F1-Score to select accurate models. XGBoost resulted with higher accuracies for prediction of career suitability, initial salary, career and salary progression with values of 92.31, 90.35, 86.45 and 88.76 respectively. Furthermore, for the prediction of professional courses and employee attrition, Random Forest resulted higher accuracies of 93.52 and 89.70. The ultimate goal of this research is to guide IT graduates and employees to have better performances and to assist them in embracing responsibilities throughout their career life.
Embargo
Machine Learning Based Solution for Improving the Efficiency of Sugar Production in Sri Lanka
(IEEE, 2022-12-26) Kulasekara, S; Kumarasiri, K; Sirimanna, T; Dissanayake, D; Karunasena, A; Pemadasa, N
Although sugar is a popularly used commodity in Sri Lanka, sugar manufactured within the country fulfill only a very small portion of the demanded amount. Sugar production is an intricate process which requires a considerable amount of expertise especially in the areas of cultivation, production and revenue prediction which may not exist in novice farmers. This research proposes a methodology which provides novice sugarcane farmers with expert knowledge on four main areas related to farming including weather forecast, sugarcane maturity estimation, production forecast and prediction of return sugarcane amounts from lands. ARIMA model is used for weather forecast whereas machine learning methods and multiple regression models were used for sugarcane maturity estimation and production of forecasts and returns respectively. The final ARIMA time series model was validated with p-value greater than 0.05 for Ljung-Box test with three different lag values. The Support Vector Machines model was identified as the best model with an accuracy of 81.19% for the sugarcane maturity estimation. The SVM model was trained using the HSV and texture features extracted from sugarcane stalk images using image processing techniques. The prediction of sugar production received a testing R-squared score of 87.75% and mean squared error of 0. Prediction of yield received a mean squared error of approximately 0 and R squared score of 98% on test data. The methodology used in this research could be used by novice farmers to increase their cultivation as well as sugar production.
Embargo
Moderate Automobile Accident Claim Process Automation Using Machine Learning
(IEEE, 2021-01-27) Imaam, F; Subasinghe, A; Kasthuriarachchi, H; Fernando, S; Haddela, P. S; Pemadasa, N
In modern-day, traditional automobile accident claim process struggles to keep up with the recurring automobile accidents and furthermore, the claim itself is a critical point in which the policyholder may decide to switch to a different automobile insurance provider. In this paper, the authors present a system which can be used to automate the processing of claims for automobiles which were involved in less severe accidents in a much quicker manner. The presented system comprises of four components, each with a model developed using computer vision or machine learning techniques to facilitate the automation process. The models are built and fine-tuned using transfer learning and ensemble learning techniques in order to determine the damaged component of the automobile, determine the make and model of the automobile, compute an accurate repair estimate and also compute the likeliness of the policyholder may churn, to ensure that the policyholder is satisfied with the appraised amount and will be retained by the insurance provider.
Embargo
Using CNNs RNNs and Machine Learning Algorithms for Real-time Crime Prediction
(IEEE, 2019-12-05) Rajapakshe, C; Balasooriya, S; Dayarathna, H; Ranaweera, N; Walgampaya, N; Pemadasa, N
Over the recent years crime rates in Sri Lanka have drastically increased. Main priority of police is to prevent crime occurrences in order to enhance public safety. Criminals use advanced technologies, which make the crime investigations cumbersome. Police officers spend lot of time and effort on these investigations. A wide range of researches are being conducted in the areas of Artificial Intelligence (AI) and Neural Networks to automate crime detection and prediction. In this paper, we present machine learning and deep learning based E-police system to enhance public safety and support law enforcement. Main objective of the system is prevention of crimes. E-Police is an application that helps police officers to get informed about the incidents happening around in real-time. In addition, system provides predictions about possible crimes likely to take place in future so that precautions can be taken to prevent those.
Embargo
UveaTrack: Uveitis Eye Disease Prediction and Detection with Vision Function Calculation and Risk Analysis Publisher: IEEE Cite This PDF
(Institute of Electrical and Electronics Engineers, 2022-10-15) Perera, B. D. K; Wickramarathna, W.A.A.I.; Chandrasiri, S; Wanniarachchi, W.A.P.W; Dilshani, S.H.N; Pemadasa, N
Uveitis is an inflammatory infection that affects uvea tissue, the middle layer of the eyewall. It can result in swelling or damage to the eye and lead to vision impairments or blindness. Most Uveitis symptoms are associated with many other diseases localized to the eye. Thus, it is hard to determine the responsible symptoms for uveitis. Consequently, early detection of this disease can prevent a perilous situation in the future. The initial motivation behind the design of this mobile application is to help accurately diagnose uveitis with minimal time and effort and thereby minimize the shortage of human specialists in this field. The 'UveaTrack' is a hybrid mobile application that enables the keep tracking of uveitis eye illness and uses machine learning (ML) algorithms, deep learning (DL) architectures, and image processing techniques for developing the system. The 'UveaTrack' application could be able to achieve an average accuracy of more than 85% and had produced overall better results. Furthermore, the 'UveaTrack' application can use as a valuable instructional tool for freshly graduated clinicians, supporting their work with patients and assisting them in making diagnostics conclusions.