Browsing by Author "Amarasinghe, T"

Now showing 1 - 2 of 2

item: Thesis-Abstract
Analysis of early dropping mechanism for optical burst switched networks
Amarasinghe, T; Thilakumara, P
Quality of Service in Optical Burst Switched Networks using Early Dropping Mechanism with Different Network Characteristics Keywords: Optical Burst Switching (OBS), Absolute Quality of Service (QoS), Early Dropping Mechanism, Dense Wave Length Division Multiplexing (DWDM). Optical Burst Switching is a promising bufferless DWDM switching technology that can potentially provide high wavelength utilization. Quality of Service support has become an important issue in OBS networks. There are two models to guarantee QoS in OBS networks. Those are relative QoS guarantee and absolute QoS guarantee. Most existing schemes are based on relative QoS model and in those models the service levels can be defined relative to the service requirements of another class of traffic. In absolute QoS model it provides a bound for loss probability of the guaranteed traffic. This kind of hard guarantee is essential to support applications with bandwidth constraints. Further efficient admission control and recourse provisioning mechanisms will enhance the service of absolute QoS model to guarantee the service requirements in the OBS networks. Early dropping mechanism is proposed to maintain the dropping probability in Absolute QoS model in OBS networks. Due to the bufferless nature of the OBS core nodes, the early dropping mechanism computes the intentional dropping probability based on measured , online loss probability. In early dropping mechanism it can be simply implemented by using a threshold value which is responsible to maintain the maximum acceptable loss probability. But in this mechanism the lower priority class of traffic suffers from high loss probability when higher priority Glasses exceed its threshold vales of loss probability. Early dropping by Span mechanism introduces a span of acceptable loss probabilities rather than using one threshold value andthis mechanism has improved QoS guarantee in higher priority classes of traffic while reducing the loss probability of lower priority classes as well. Further the performance of this mechanism can be applied in a dynamic wave length assigning network in order to guarantee the absolute QoS with efficient recourse provisioning
item: Conference-Full-text
Machine learning-based automated tool to detect Sinhala hate speech in images
(Faculty of Information Technology, University of Moratuwa., 2021-12) Silva, E; Nandathilaka, M; Dalugoda, S; Amarasinghe, T; Ahangama, S; Weerasuriya, GT; Ganegoda, GU; Mahadewa, KT
Social media platforms have emerged rapidly with technological advancements. Facebook, the most widely used social media platform has been the primary reason for the spread of hatred in Sri Lanka in the recent past. When a post with Sinhala hate content is reported on Facebook, it is translated to the English language before the review of the moderators. In most instances, the translated content has a different context compared to the original post. This results in concluding that the reported post does not violate the established policies and guidelines concerning hate content. Hence, an effective approach needs to be in place to address the aforementioned problem. This research project proposes a solution through an automated tool that is capable of detecting hate content presented in Sinhala phrases extracted from Facebook posts/memes. The tool accepts an image that contains Sinhala texts, extracts the text using a Convolutional Neural Network (CNN) model, preprocesses the text using Natural Language Processing (NLP) techniques, analyzes the preprocessed text to identify hate intensity level and finally classifies the text into four main domains named Political, Race, Religion and Gender using a text classification model.