Research Papers - Dept of Software Engineering
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/1022
Browse
3 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Embargo Optimization of Volume & Brightness of Android Smartphone through Clustering & Reinforcement Learning (“RE-IN”)(IEEE, 2018-12-21) Abeywardhane, J. S. D. M. D. S; de Silva, E. M. W. N; Gallanga, I. G. A. G. S; Rathnayake, L. N; Wickramaratne, C. J; Sriyaratna, DSmartphone has become one of the most significant piece of technology that humans were able to produce in the 21st century. It has become our life companion; hence the features of the smartphones have developed in advance. But, some features may not work as expected. For instance, auto brightness changing feature is now actualized with smartphones, yet we alter the brightness according to our preference. In the same manner, considering the volume of our smartphone it doesn't change according to our preference subsequently. This research will develop a mobile application (“RE-IN”) to overcome this issue for Android smartphones. Since android smartphones allow accessing its hardware layer we can roll out improvements as we need, yet Apple doesn't permit to proceed with its hardware layer thus hard to do this for the iPhone users. By utilizing the RE-IN mobile application users may have to encounter an optimal brightness and volume on their Android smartphones agreeing the present condition of smartphone users are in. RE- IN application will keep running as a background application on an Android smartphone. When the client changes the brightness and volume as his/her preference. At that point, the reinforcement learning algorithm over the time application will distinguish how to control user's smartphone's brightness and volume relying upon the user's circumstance. When client surrounding is loaded with light, the framework will modify brightness for his/her preference. The client doesn't need to do this manually. Moreover when the client is at the too much boisterous place all of a sudden gets a call from someone; client's smartphone amplifier volume will change consequently and solaces the client's discussion. To actualize this framework it is relied upon to reinforcement learning and machine learning as the research area. By finishing the literature review, research group unable to find an Android mobile application which automates the process of volume and brightness of the Android smartphone as per user preference. After using the reinforcement learning algorithm to learn the data set then distribute the process, using client-server model and come up with a clustering algorithm(K-means algorithm) to share common attributes by considering geographical area which they live in and variables like age, gender, how they interact with the device etc. In addition, this system will identify abnormal behaviors of some particular users. RE-IN will identify the users who are keeping volume level to the highest and brightness level to its maximum and notify them in advance.Publication Embargo Predicting Diabetes Mellitus Using Machine Learning and Optical Character Recognition(IEEE, 2021-04-02) Silva, W. A. J. R; Shirantha, H. M. K; Balalla, L. J. M. V. N; Ranasinghe, R. A. D. V. K; Kuruwitaarachchi, N; Kasthurirathna, DDiabetes Mellitus is recognized as a chronic metabolic disease that is characterized by hyperglycemia. As stated by the International Diabetes Federation, the statistics reveal that the incidence of diabetes among adults in Sri Lanka is 8.5%. In hindsight, this indicates that an average of one in every twelve adults in Sri Lanka is at risk of being diagnosed with the disease. However, presently, due to the lack of knowledge or mediums concerning the disease and its symptoms, diabetes often goes undetected which has resulted in 1/3 rd of the constituent population being unaware that they possess the disease. The proposed system aims to implement an application to read and analyze medical reports which will generate data that predicts the probabilities of the contraction and onset of diabetes, with insurance of maximum system efficiency and data credibility. Machine learning classification algorithms and optimization techniques have been used to predict diabetes status with maximum accuracy. To extract data from medical reports Optical Character Recognition, Image Processing, and Natural Language Processing have been usedPublication Embargo Optimization of Microservices Security(IEEE, 2021-12-09) Kalubowila, D. C; Athukorala, S. M; Tharaka, B. A. S; Samarasekara, H. W. Y. R; Samaratunge Arachchilage, U. S. S; Kasthurirathna, DMicroservices is a trending architecture, and due to its demanding features and behaviors, billions of business applications are developed based on it. Due to its remarkable ability to deploy and coordinate containerized microservices, Kubernetes deployments support the service mesh architectures, and that ensures secured inter-service communication. The Istio is the widely used service mesh tool at present. However, service-to-service communication happens in the present Istio architecture, and there is a probability of exchanging unauthorized and over-provisioned requests due to incorrect implementation. Currently, these requests are verified within the upstream microservice. Obtaining a response to an erroneous request may take considerable time latency. This research thereby aims to address a solution to reduce the latency of a response by implementing an external validation model. The proposed external validation model ensures that the required parameters are validated and actions are taken before requests reach the service level. External validation enables applications to save significant time and resources.