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

Filters

2011 - 201922020 - 20252

Advanced Search

Filter by

Settings

Has files: YesSubject: search.filters.subject.neural networks

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    Evaluation of Machine Learning Models in Student Academic Performance Prediction
    (Institute of Electrical and Electronics Engineers Inc., 2025) Sandeepa A.G.R.; Mohottala, S
    This research investigates the use of machine learning methods to forecast students' academic performance in a school setting. Students' data with behavioral, academic, and demographic details were used in implementations with standard classical machine learning models including multi-layer perceptron classifier (MLPC). MLPC obtained 86.46% maximum accuracy for test set across all implementations while for train set, it was 99.45%. Under 10-fold cross validation, MLPC obtained 79.58% average accuracy for test set while for train set, it was 99.65%. MLP's better performance over other machine learning models strongly suggest the potential use of neural networks as data-efficient models. Feature selection approach played a crucial role in improving the performance and multiple evaluation approaches were used in order to compare with existing literature. Explainable machine learning methods were utilized to demystify the black box models and to validate the feature selection approach.
  • Thumbnail Image
    PublicationOpen Access
    Neural Network based automated hot water mixture
    (SLIIT, Faculty of Engineering, 2023-03-02) Firsan, F.N.M; Herath, G.M; Thilakanayake, T.D
    In the present day and age, most residential spaces comprise a shower system and generally a conventional system of hot water showers. Throughout history, showering has developed as an essential need in a person’s life. Nevertheless, a typical hot water shower system comprises delays in hot water mixing and usually requires an average of 2 to 4 minutes to mix the cold and hot water to deliver the appropriate shower temperature. The delay in mixing provides less comfort and poor satisfaction affecting people’s lifestyles. Due to these disadvantages, a system incorporating artificial Intelligence can be utilized to enhance the performance of mixing which can offer an automated hot water mixture system with improved efficiency and effectiveness. Recently, significant research has been focused on utilizing deep learning technology due to its multiple breakthroughs in fabricating a broad range of automated novel applications since Neural Networks comprise the capacity to learn from data to offer efficient and accurate systems. In this research project, the hot water mixture is employed by an Artificial Neural Network model integrated with the combination of an embedded system of the proposed system of hot water mixture. Furthermore, the proposed system comprises temperature and flow sensors along with controllable flow valves. The tested system indicated acceptable accuracy between the actual and desired output flow rate and temperature.
  • Thumbnail Image
    PublicationEmbargo
    An adaptive routing algorithm for Cognitive Packet Network infrastructure based on neural networks
    (IEEE, 2011-08-16) Madubashitha, D. K. D; Wijesinghe, W. M. S. S; Kamaladiwela, K. A. S. R; Ranaweera, M. G. P; Wijekoon, J; Abeygunawardhana, P. K. W
    This paper examines the possibility of introducing an intelligent routing protocol to the Internet, based on the Cognitive Packet Network (CPN) architecture with respect to the Quality of Service (QoS) delivered to the end users. In the present with increasing populations of countries it is clear that present infrastructure does not hold the sufficient capacity to deliver the expected level of service to the end users. Since there is an eminent need for a solution for improving the QoS in the Internet, this research focuses to provide a new network architecture which would improve the QoS, provide reliable and efficient service which can fulfill the ever growing Internet usage demand. This is achieved through a new network architecture known as CPN which is based on the basis of providing the best and user desired QoS. The main underlying technology behind the CPN will be a neural network. The neural network will be learning the changes in the network and adapt to the situation through the knowledge gathered. The packets will collectively learn about the network thus the load on the routers will be minimized. This mechanism completely replaces the need of a routing table thus making routing far more efficient when comparing to current routing protocols like Open Shortest Path First (OSPF). Final outcome of the research is coming to the conclusion that the future of the Internet is with the neural network based intelligent, dynamically adapting and learning CPN infrastructure instead of current packet switched network.
  • Thumbnail Image
    PublicationEmbargo
    An adaptive routing algorithm for Cognitive Packet Network infrastructure based on neural networks
    (IEEE, 2011-08-16) Madubashitha, D. K. D; Wijesinghe, W. M. S. S; Kamaladiwela, K. A. S. R; Ranaweera, M. G. P; Wijekoon, J; Abeygunawardhana, P. K. W
    This paper examines the possibility of introducing an intelligent routing protocol to the Internet, based on the Cognitive Packet Network (CPN) architecture with respect to the Quality of Service (QoS) delivered to the end users. In the present with increasing populations of countries it is clear that present infrastructure does not hold the sufficient capacity to deliver the expected level of service to the end users. Since there is an eminent need for a solution for improving the QoS in the Internet, this research focuses to provide a new network architecture which would improve the QoS, provide reliable and efficient service which can fulfill the ever growing Internet usage demand. This is achieved through a new network architecture known as CPN which is based on the basis of providing the best and user desired QoS. The main underlying technology behind the CPN will be a neural network. The neural network will be learning the changes in the network and adapt to the situation through the knowledge gathered. The packets will collectively learn about the network thus the load on the routers will be minimized. This mechanism completely replaces the need of a routing table thus making routing far more efficient when comparing to current routing protocols like Open Shortest Path First (OSPF). Final outcome of the research is coming to the conclusion that the future of the Internet is with the neural network based intelligent, dynamically adapting and learning CPN infrastructure instead of current packet switched network.