1st International Conference on Advancements in Computing [ICAC] 2019
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/1599
Browse
68 results
Search Results
Publication Embargo Traffic Density Estimation and Traffic Control using Convolutional Neural Network(IEEE, 2019-12-05) Ikiriwatte, A. K; Perera, D. D. R; Samarakoon, S. M. M. C; Dissanayake, D. M. W. C. B; Rupasignhe, P. LThe existing traffic light control systems are inefficient due to the usage of predefined algorithms on offline data. This causes in numerous problems such as long delays and a wastage of energy. Estimation of traffic density indirectly affects in decreasing the high traffic congestion which will occur due to the less planning of transportation infrastructure and the policies. The goal of this research is to introduce an applicable method to improve the existing static traffic signal system into a dynamic system. As an approach we analyze the use of machine learning algorithms to measure the traffic density to tackle this research problem of high traffic congestion. The main target is to implement this system for the four-way junctions since it is a place where the possibility of having a traffic congestion seems to be high. With use of these traffic density estimation algorithm, crowd density estimation and signal handling we conduct experiments on minimizing the congestion at four-way junctions. We decided on using convolutional neural networks as an advanced machine learning method to increase the accuracy of the learning algorithm.Publication Embargo Secure Web Navigation with Intrusion Detection And Quota Management for SOHO and Small Scale Businesses(IEEE, 2019-12-05) Perera, M. A. D. S. R; Hemapala, C; Udugahapattuwa, M; Senarathne, A. NIt's a modern day necessity and a trend to offer free and open web access to their customers and employees in small scale and Small Office Home Office (SOHO) business culture (restaurants, malls, coffee shops). Unfortunately, internet security and quota management are mostly overlooked which makes it an intruders' paradise. The existing solutions that incorporate machine learning based dynamic aspects, cannot be afforded by our target audience nor do they possess the extensive IT knowledge to configure and maintain them. To cater to this gap, this research proposes the network management device `Dynamic Defender', targeted for small scale and SOHO type businesses which focuses on affordability and user-friendliness as key factors while incorporating cutting edge machine learning technologies. The Dynamic Defender's Intrusion Detection Engine is comprised of 99.13% accuracy with its base run on Artificial Neural Networks. URL Classification Engine produced high accuracy on all 3 machine learning algorithms which were used. Specifically, Random Forest with 92.94 % precision, Artificial Neural Networks with 90.33% precision and Logistic Regression with 91.41% precision. The Dynamic Bandwidth Management System consisted of an 89% accuracy level on the hybrid module of Linear Regression and Decision Trees while the Quota Management System (QMS) provided an accuracy level of 82% in K-Nearest Neighbors and 89% on Decision Tree algorithm.Publication Embargo Security Platform for Mobile OS(IEEE, 2019-12-05) Benett, A. S. B; Vinushanth, K; Ranjitha, L; Abisherk, R. S; Senarathne, A. NEvolution of human is evident in everything that we see, feel and use today. History of phone is one such example we can see. Modern devices have all the features that helped it to become the ultimate source of data for an individual. It was easy for an individual to keep all his data intact with him in his hands. There are vulnerable points which can be exploited to acquire the personal and sensitive data from the device in order to gain unethical advantage over an individual. Bluetooth, Wi-Fi and human errors are some of those vulnerable points. In this paper, multilevel malware detection with the help of machine learning, Rogue access point detection and accidental data leakage prevention are proposed with an emphasis on Android mobile operating system. As a result, accidental leakage of sensitive data by the user can be prevented. Further, rogue access point detection will help the user to prevent data loss through wireless network and the malware detection can prevent all the known and 85% of the unknown malwares.Publication Embargo Enhanced Secure Solution for PoS Architecture(IEEE, 2019-12-05) Samaranayake, C; Kuruppu Achchige, R. P; Shanaz, T; Ranasinghe, A; Senarathne, A. NToday retail businesses expect to bring the utmost in sales and payment transactions by adapting new technologies. Therefore, Advanced Point of Sales (PoS) Systems are widely used in the industry. Regardless of how efficient and secure these systems or applications work, unexpected information security risks can arise. Such risks could be a threat to their business and organization. It is important to ensure that critical information such as payment card information, handled in PoS systems is kept secure from attacks that could bring financial loss. This research provides a solution by studying the overall infrastructure of a PoS System and identifies the key events that such data would be at risk. The major concern of it was to enhance the existing security features of the system to avoid any type of malicious activity. This research consists of four main sections under security related to PoS Systems that would address the risk; Studying of malware and classifying them, detecting possible attacks and means of preventing it, a robot (BOT) to predict and generate the system status with a Data Leakage Prevention(DLP) solution for all the events occurring at a PoS. The key objective of implementing this solution was to protect the confidential data that is being used in the PoS System and to avoid threats that lead to the unavailability of the system. The implemented security features using machine learning and Deep Learning methods to the existing PoS functions produced a 99.3% of accuracy in Malware Detection and 95% of accuracy in its Classification process while the DLP Solution was able to obtain an accuracy of 84.6%. The above results retrieved fulfilled the research objectives and aided to integrate an enhanced security solution for a PoS system.Publication Embargo Optimization of Customer-Friendly Manual Load Shedding System(IEEE, 2019-12-05) Fernando, W. D. I; Rankothge, W; Perera, A. D. S; Dissanayake, S. J; De Silva, W. D. STo maintain the supply and demand of electricity power, load-Shedding is one of the methods practiced by the energy suppliers to hold the power system balanced, when an energy deficit problem arises. Lacking a proper load shedding scheme will lead to system instability and it will cause serious system frequency decay. We have proposed a solution to optimize the manual load shedding schedule with the application of optimization techniques, specifically the Genetic Algorithms. We have considered current hold by all feeders throughout the country, and the time period of load shedding as main factors in the optimization model. Our results show that, using our proposed model, we can minimize the imbalance between the supply and demand of electricity by selecting the best feeder to be selected for load shedding under given constraints.Publication Embargo Multi-Cloud Container Communication Using Software Defined Networks(IEEE, 2019-12-05) Najath, M; M.S Fayas Akram, M. F. S; Ahamed, I. A; Rankothge, WOver the last decade, cloud computing has changed the IT world, mainly in the ways of storing, accessing and securing the users' data. However, there is no single cloud management platform in the market that can be used to manage different clouds (ex: AWS and AZURE), together with container communication between the clouds, to cater different cloud user requirements. We have proposed a platform for cloud service providers that automates the multi-cloud management related services with following modules: (1) communication, (2) resource allocation and (3) log view. Our results show that, using our proposed modules, the multi-cloud management related services can be automated efficiently and reliably.Publication Embargo VAULT - A Shared Distributed And Redundant Storage Solution(IEEE, 2019-12-05) Peiris, T. R. N. R; Bandara, W. M. U. K. M. T; Sachintha, K. V. A; Senarathne, A. NAn ideal distributed storage solution must have the ability to provide redundant, reliable, shared and secure access to user data without compromising the ability to scale and descend while maintaining performance. VAULT is an attempt to avert the negatives of the cloud in a local environment using a decentralized methodology. VAULT makes use of individual idle storage space on a network of peer-to-peer nodes which is then provided to an end user to store files in the pooled space. VAULT implements redundancy by the use of Reed-Solomon codes and maps file fragment locations using a blockchain as a distributed ledger. Fragment distribution is optimized using a machine learning approach where node characteristics are used to determine the reliability of each node. The aggregation of above features makes VAULT an ideal solution for corporate environments where consumer hardware and infrastructure is already allocated.Publication Embargo Recognition and translation of Ancient Brahmi Letters using deep learning and NLP(IEEE, 2019-12) Wijerathna, K. A. S. A. N; Sepalitha, R; Thuiyadura, I; Athauda, H; Suranjini, P. D; Silva, J. A. D. C; Jayakodi, AInscriptions are major resources for studying the ancient history and culture of civilization in any country. Analyzing, recognizing and translating the ancient letters (Brahmi letters) from the inscription is a very difficult work for present generation. There is no any automatic system for translating Brahmi letters to Sinhala language. However, they are using manual method for translating inscriptions. The method that used in epigraphy is being taken a long period to decipher, analyze and translate the inscribed text in inscriptions. This research mainly focuses on recognition of ancient Brahmi characters written the time period between 3 rd B.C and 1 st A. D. First, we remove the noise, segment the letters from the inscription image and convert it into the binary image using image processing techniques. Secondly, we recognize the correct Brahmi letters, broken letters and then identify the time period of the inscriptions using Convolution Neural Networks in deep learning. Finally, the Brahmi letters are translated into modern Sinhala letters and provide the meaning of the inscription using Natural Language Processing. This proposed system builds up solution to overcome the existing problems in epigraphy.Publication Embargo Easy Learning: Augmented Reality Based Environmental Studies for Primary Students(IEEE, 2019-12-05) Wickramapala, T; Jayawardhana, L; Tharaki, S; Senevirathna, S; Gamage, N; Wickramarathna, JPrimary education is every child's fundamental right. According to the United Nations Educational, Scientific and Cultural Organization (UNESCO), primary schooling offers learning and educational activities typically intended to provide learners with basic abilities in reading and writing. The students find it difficult to identify trees and animals around them due to the lack of exposure to the natural environment. This research study introduces mobile based application (Easy Learning) which embraced augmented reality technology (AR) to motivate and aid learners in studying Environmental Studies in terms of identification of animals and trees. In order to provide sufficient knowledge about trees and animals, this research focuses on safe internet browsing and summarization for trees and animals. Easy learning suggest safe videos for kids and generates knowledge based questions to evaluate themselves as well. The study also evaluates whether the students like the features of the Easy Learning and the rate of knowledge change, through pre and post questionnaires given at the beginning and at the end of the implementation of Easy Learning. The findings proves that the Easy Learning as an interactive AR based learning instrument, for Environmental Studies which improves the learning curve.Publication Embargo Smart Platform for Cloud Service Providers(IEEE, 2019-12-05) Dharmapriya, W. A. S. P; Supipi, K. G; Ravindu Nimesh, G. G; Muhandiram, M. A. B. K; Rankothge, W. H; Gamage, NCloud computing offers many types of computer related services without the direct active management of their users. Cloud Service Providers (CSPs) are responsible to manage these services such as placement of services in the cloud, resource allocation, network monitoring etc. The cloud service provider is required to monitor the network traffic, predict the dynamic traffic changes, and scale out the resources accordingly. We have proposed a platform for cloud service providers that automates the cloud management related services with following modules: (1) traffic monitoring, (2) traffic prediction, (3) virtual service instances placement and (4) traffic load balancing. We have used continuous and periodic approaches for traffic monitoring, Auto-Regressive Integrated Moving Average (ARIMA) model for traffic prediction, Randomized Weighted Majority Algorithm (RWMA) for virtual service instances placement and a threshold-based approach for load balancing. In this paper, we are presenting the performances of our cloud management platform, specially an evaluation of the algorithms used in above mentioned modules. Our results show that, using our proposed modules, the cloud management related services can be automated efficiently and reliably.