Papers by B. T. G. S. Kumara
CNN Based Severity Prediction of Bug Reports
2021 From Innovation To Impact (FITI)
Docker Containerized Infrastructure Orchestration with Portainer Container-native Approach
2022 3rd International Conference for Emerging Technology (INCET)
Performance Evaluation of Docker-based Apache and Nginx Web Server
2022 3rd International Conference for Emerging Technology (INCET)
TRETA - A Novel Heuristic Based Efficient Task Scheduling Algorithm in Cloud Environment
2020 IEEE REGION 10 CONFERENCE (TENCON)
Cloud computing is a computing platform that allows users to access various kinds of computing se... more Cloud computing is a computing platform that allows users to access various kinds of computing services over the internet. Cloud provides on-demand, scalable and highly available resources on pay-per-usage subscriptions. Cloud is an optimum solution for executing a large number of different size tasks as for the computing capability it offers. Task scheduling is one of the major open challenges that need to be addressed. The Task scheduling problem in the cloud is known to be an NP- complete problem. Hence heuristics can be used to get an optimal solution. There have been many heuristics proposed for the task scheduling problem in the cloud. None of them has considered the total execution time of the virtual machine as a factor for finding a better schedule. In this paper, we proposed a new task scheduling algorithm named Total Resource Execution Time Aware Algorithm (TRETA) which takes into account the total execution time of computing resources in obtaining an optimal schedule. The algorithm is compared with Min-Min, Min-Max, FCFS, and MCT heuristics for Makespan, Degree of Imbalance and System Throughput. The proposed algorithm shows a significant amount of improvement in Makespan compared to other heuristics. The algorithm also outperforms other heuristics with respect to System Throughput and Degree of Imbalance which results in better workload distribution among the cloud resources.
An Efficient Task Scheduling Algorithm using Total Resource Execution Time Aware Algorithm in Cloud Computing
2020 IEEE International Conference on Smart Cloud (SmartCloud)
Cloud computing is a new computing paradigm that allow users to access services over the internet... more Cloud computing is a new computing paradigm that allow users to access services over the internet. Cloud provides scalable, highly available and on-demand resources to its endusers. The cloud charges its customers for only the usage of the services. The cloud platform offers a range of computing capabilities with numerous configurable variations. Hence cloud can be used to execute a large number of tasks with different computing and data requirements. To benefit from the cloud and get the optimum, efficient management of the underlying system is essential. Task scheduling is an open problem that needs to be countered and it is known to be an NP-complete problem. Therefore heuristic techniques can be used to derive a better solution for scheduling tasks. Several heuristics has been proposed for addressing the task scheduling problem in the cloud environment considering many factors in finding a better schedule. None of them has considered the total execution time of the virtual machine as a factor for finding a better schedule. In this paper, we propose, Total Resource Execution Time Aware Algorithm (TRETA). The algorithm considers the total execution time of computing resource in finding an optimal schedule. The algorithm is compared with existing state-of-art heuristics Min-Min, Min-Max, FCFS, and MCT for Makespan, Degree of Imbalance, and System Throughput using real-world workload traces of Nasa Ames iPSC/860. The experimentation is performed for workloads consisting of a bigger set of tasks using CloudSim 5.0 Simulator. The proposed algorithm shows a significant amount of improvement in Makespan, Degree of imbalance, and System Throughput compared to other existing heuristics.
Deep Learning Approach for Detecting Customer Churn in Telecommunication Industry
Social Customer Relationship Management (Social-CRM) in the Era of Web 4.0
In today's business world, customer turnover is a significant problem. Communications compani... more In today's business world, customer turnover is a significant problem. Communications companies aren't exempt from these problems. Retaining consumers is more important than recruiting new ones when it comes to business. Getting new clients is about five times as expensive as keeping old ones in this field. As a result, anticipating client turnover is a huge challenge for almost all organizations. This study focused on analyzing information on around 7000 post-paid subscribers by considering 21 different attributes. Initially, the data was fed into machine learning techniques such k-nearest neighbors, artificial neural networks, etc. In addition, deep neural networks (DNN) have also considered more than one hidden layer. A total of 4284 of the 7234 post-paid customers are considered non-churners, while the remaining 2950 are churners. The long short-term memory networks (LSTM) considered under the DNN produce results far superior to the other techniques, with the highest acc...
LSTM Based Approach for Classifying Twitter Posts for Movie Success Prediction
2020 International Conference on Decision Aid Sciences and Application (DASA)
Social media like Twitter contains rich information about people's preferences. There is a st... more Social media like Twitter contains rich information about people's preferences. There is a struggle to determine how to effectively utilize and interpret those data. Models are created using these large quantities of data to predict the behavior and tendencies. People share their thoughts about movies on Twitter. The movie industry has been a very important sector in the global market. So, it is important to maximize the profit by predicting the success of the movie. In this study, we proposed an approach to classify twitter posts predict the success of movies using Long short-term memory based approach. Our approach outperformed the existing approach by obtaining 83.97% accuracy.
Automatic Classification of Questions based on Bloom's Taxonomy using Artificial Neural Network
2022 International Conference on Decision Aid Sciences and Applications (DASA)
A Machine Learning Approach to Classify the Telecommunication Customers Based on Their Profitability
Social Customer Relationship Management (Social-CRM) in the Era of Web 4.0
Customer profitability is one of the most critical problems faced by businesses today. Keeping an... more Customer profitability is one of the most critical problems faced by businesses today. Keeping an existing customer is more valuable than gaining a new subscriber in the telecommunication industry. As a result, anticipating customer attrition behavior in advance is challenging. This behavior has prompted most researchers to establish a model for categorizing clients based on their profitability levels in various businesses. This study was carried out with the assistance of a local telecommunication service provider. Approximately 10,000 pre-paid subscriber details with 12 attributes were acquired. Furthermore, the classification technique was used to reduce the dimensionality between features and classify the high profitable customers, low profitable customers, and average profitable customers. The data was then fed into various supervised learning algorithms to choose the optimum algorithm by considering certain evaluation metrics for developing the final prediction model. The prop...
Classify the Outcome of Arterial Blood Gas Test to Detect the Respiratory Failure Using Machine Learning
2022 International Conference on Decision Aid Sciences and Applications (DASA)
Work From Home After Covid-19: Machine Learning-Based Approach to Predict Employee's Choice
2022 International Conference on Decision Aid Sciences and Applications (DASA)
Well before the Covid-19 pandemic, Work from Home (WFH) had become an increasingly common practic... more Well before the Covid-19 pandemic, Work from Home (WFH) had become an increasingly common practice among employees. With the pandemic outbreak, millions of employees across the globe were forced to shift into full-time WFH in response to continuous lockdowns or health restrictions. It is expected that a considerable number of organisations will continue the WFH practice in a hybrid mode along with Work From Office (WFO) even after the pandemic. This study aims to predict employees' choice on continuing WFH after the pandemic. The data set was collected using an online questionnaire shared among a sample of employees engaged in WFH during the pandemic. Naïve Bayes, Artificial Neural Network (ANN), Random Forest, and Ensemble Learning-based approaches were used to generate the prediction models. Ensemble Learning-based approach was the best classifier compared to the other three classifiers, and it obtained a 91.6% accuracy value. Naïve Bayes showed the lowest performance
Machine Learning Approach to Detect Online Shopping Addiction
2022 2nd International Conference on Advanced Research in Computing (ICARC)
Online shopping addiction is the rapidly growing phenomenon. The excessive or uncontrollable buyi... more Online shopping addiction is the rapidly growing phenomenon. The excessive or uncontrollable buying over the internet can be defined as online shopping addiction. Compulsive buying disorder is the mental health problem that associated with shopping addiction. This addictive situation directly affected to the consumers and caused to several negative impacts. This study focuses on detect online shopping addiction by considering consumers’ motivation towards attractive features and facilities provided by the online shopping environment. The data were collected through online questionnaire and 511 data were used for carry out this research. The Multilayer Perceptron (MLP) neural network, SVM, Naïve Bayes, Random Forest and Decision Tree algorithms were used to develop the models. For the neural network model there were 11 attributes in the input layer and 2 classes in the output layer. Then add one hidden layer with 13 neurons to train and build the model. The accuracy of the build MLP model was 90.90% and it was the highest accuracy compared to the other developed models. This machine learning model can find out the addiction towards online shopping.
Adoption of Docker Containers as an Infrastructure for Deploying Software Applications: A Review
Advances on Smart and Soft Computing, 2021
TRETA-A Novel Heuristic Based Workflow Scheduling Algorithm in Cloud Environment
2020 IEEE 15th International Conference on Industrial and Information Systems (ICIIS), 2020
Cloud computing is a new computing paradigm that lets users access services over the internet. Cl... more Cloud computing is a new computing paradigm that lets users access services over the internet. Cloud provides scalable, on-demand resources and highly accessible. Cloud charges its customers for only the usage. Workflow applications are used in many business processing, scientific fields and in many other domains. Cloud has become one of the optimum solutions for executing workflows as for the computing power and the benefits it offers. Workflow scheduling can reduce the overall cost of execution and optimize resource utilization in the cloud for both the cloud consumer and the service provider. In this paper, We compare our novel algorithm Total Resource Execution Time Aware Scheduling Algorithm (TRETA) with existing heuristics which considers the total execution time of the computing resource as a factor for finding an optimal schedule. To the best of our knowledge none of the previous work has not considered the above metric for finding a better schedule. We compare the proposed ...
2019 International Research Conference on Smart Computing and Systems Engineering (SCSE), 2019
Anthropometric measurements are generally used to determine and predict achievement in different ... more Anthropometric measurements are generally used to determine and predict achievement in different sports. An athlete's anthropometric and physical characteristics may perform important preconditions for successful participation in any given sport. Further, anthropometric profiles indicate whether the player would be suitable for the competition at the highest level in a specific sport. Recently, more researches have been carried out on Sport Data mining. In this study, we propose an approach to identify the most suitable sport for beginners using data mining and anthropometric profiles. We propose clustering base approach. We apply a spatial clustering technique called the Spherical Associated Keyword Space which is projected clustering result from a three-dimensional sphere to a two dimensional (2D) spherical surface for 2D visualization. Empirical study of our approach has proved the effectiveness of clustering results.
Docker incorporation is different from other computer system infrastructures: A review
2021 International Research Conference on Smart Computing and Systems Engineering (SCSE), 2021
Currently the computing world is getting complex, innovating and maturing with modern technologie... more Currently the computing world is getting complex, innovating and maturing with modern technologies. Virtualization is one of the old concepts and currently containerization has arrived as an alternative and innovative technology. Docker is the most famous and trending container management technology. Different other container management technologies and virtualization technologies are respective other corresponding technologies and mechanisms for Docker containerization. This research study aims to identify how Docker incorporation is different from other computer system infrastructure technologies in the perspective of architecture, features and qualities. By considering forty-five existing literatures, this research study was conducted. To deliver a structured review process, a thorough review protocol was conducted. By considering four main research questions, the research study was lined up. Ultimately, Docker architecture and Docker components, Docker features, Docker integration with other computing domains and Docker & other computing infrastructures were studied. After synthesizing all the selected research studies, the cream was obtained with plenty of knowledge contribution to the field of computer application deployment and infrastructure.
2019 International Research Conference on Smart Computing and Systems Engineering (SCSE), 2019
Haphazard development activities on mountain slopes and inadequate attention to construction aspe... more Haphazard development activities on mountain slopes and inadequate attention to construction aspects have led to the increase of landslides and consequently sustaining damage to lives and infrastructure. Nearly 3275 sq.km of area spread over the Ratnapura District, seems to be highly prone to landslides and mass wastage of 2178 sq.km. Landslides occur in many regions of Ratnapura district and nearly 90 deaths have been reported according to National Research Building Organization (NBRO) in 2017. Most landslides or potential failures could be predicted fairly accurately if proper investigations were performed in time. The primary objective of this study is landslide-hazard mapping and risk evaluation to determine the real extent, timing, and severity of landslide processes in Ratnapura district. Such knowledge will provide the most significant benefit to government officials, consulting engineering firms, and the general public in avoiding the landslide hazard or in mitigating the losses. Hybrid Machine Learning techniques can be used to develop prediction models using existing data. Ensemble approach based on Support Vector Machine (SVM), Naïve Bayes model were combined and implemented for the final prediction. This study possesses a strong capability to predict landslides by causative factors, slope, land use, elevation, geology, soil materials and triggering factor; rainfall was extracted and applied to the machine learning algorithms. This research introduces a novel architecture to produce a more relevant and accurate prediction of the landslide vulnerability within the study area. Moreover, it was revealed that all of the factors had relatively positive relationship with occurrence of landslides. An improvement in hazard monitoring, accuracy of early warning and disaster mitigation is documented.
Constructing Global Researchers Network Using Google Scholar Profiles for Collaborator Recommendation Systems
2021 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT), 2021
Researchers like to collaborate with other researchers to share their knowledge, experience, and ... more Researchers like to collaborate with other researchers to share their knowledge, experience, and resources with each other. By selecting appropriate collaborators researchers can obtain accurate output and they can publish more papers with high quality. They can get higher recognition within their research community. However, selecting appropriate collaborators is a challenging task. Thus, researchers proposed collaborator recommendation systems (CRS) to address that challenge. A network-based recommendation system is one type of CRS. Constructing an effective network is very important in the CRS. Existing network constructing approaches used co-author lists or publications to generate the network. However, the method is not suitable for some researchers like junior researchers and undergraduate students due to a lack of co-authors and publications. In this paper, we propose Google Scholar-based researchers network constructing approach. Common coauthors, the similarity of the area of interests, citation rate, and several co-authored publications between two researchers are extracted from GS to construct the network. The empirical work of our prototype system shows the efficacy of the technique presented.
Constructing Global Researchers Network Using Google Scholar Profiles for Collaborator Recommendation Systems
2021 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT), 2021
Researchers like to collaborate with other researchers to share their knowledge, experience, and ... more Researchers like to collaborate with other researchers to share their knowledge, experience, and resources with each other. By selecting appropriate collaborators researchers can obtain accurate output and they can publish more papers with high quality. They can get higher recognition within their research community. However, selecting appropriate collaborators is a challenging task. Thus, researchers proposed collaborator recommendation systems (CRS) to address that challenge. A network-based recommendation system is one type of CRS. Constructing an effective network is very important in the CRS. Existing network constructing approaches used co-author lists or publications to generate the network. However, the method is not suitable for some researchers like junior researchers and undergraduate students due to a lack of co-authors and publications. In this paper, we propose Google Scholar-based researchers network constructing approach. Common coauthors, the similarity of the area of interests, citation rate, and several co-authored publications between two researchers are extracted from GS to construct the network. The empirical work of our prototype system shows the efficacy of the technique presented
Work From Home After Covid-19: Machine Learning-Based Approach to Predict Employee's Choice
2022 International Conference on Decision Aid Sciences and Applications (DASA), 2022
Well before the Covid-19 pandemic, Work from Home (WFH) had become an increasingly common practic... more Well before the Covid-19 pandemic, Work from Home (WFH) had become an increasingly common practice among employees. With the pandemic outbreak, millions of employees across the globe were forced to shift into full-time WFH in response to continuous lockdowns or health restrictions. It is expected that a considerable number of organisations will continue the WFH practice in a hybrid mode along with Work From Office (WFO) even after the pandemic. This study aims to predict employees' choice on continuing WFH after the pandemic. The data set was collected using an online questionnaire shared among a sample of employees engaged in WFH during the pandemic. Naïve Bayes, Artificial Neural Network (ANN), Random Forest, and Ensemble Learning-based approaches were used to generate the prediction models. Ensemble Learning-based approach was the best classifier compared to the other three classifiers, and it obtained a 91.6% accuracy value. Naïve Bayes showed the lowest performance
Uploads
Papers by B. T. G. S. Kumara