4th International Conference on Advancements in Computing [ICAC] 2022
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/3384
Browse
Publication Embargo Real-time Smart Navigation System for Visually Impaired People(IEEE, 2022-12-09) Sudaraka Keshara, S.R.D.; Weragoda, W.R.J.M.; Chandrasiri, S; Ellankovan, J; Madushan, W.AVisual sense plays a primary role in guiding sighted people through an unfamiliar environment and assisting them to reach their destination safely. Visual impairment describes the actual damage that makes it difficult to accomplish visual tasks because it makes it difficult to see clearly. This paper proposes an approach to overcome the challenges faced by visually impaired people with the help of machine learning. This proposed system combines a smart cane and a wearable smart glass. The detection of obstacles and potholes helps to increase the safety and comfort of visually impaired users by detecting and displaying obstacles, and the Smart Walk-lane Navigation assists in navigating through the walk-lanes without letting them enter the main roads and helps to prevent accidents. Road sign detection allows users to follow road signs and cross the roads safely, while face and emotion detection allows users to recognize well-known people and their emotions.Publication Embargo Using Sentiment Analysis to Explore the Accommodation Experience in the Sharing Economy through Topic Modeling(IEEE, 2022-12-09) Bandara, H.M.A.P.; Charles, J.; Lekamge, L. S.The rapid proliferation of internet-based technology has made the sharing economy the next e-commerce business model. Recently, sharing economy lodging platforms have gained a significant market share in the tourism and lodging industry. Tourism and hospitality industries are now being significantly disrupted by Airbnb, an online lodging platform. For businesses and customers who utilize these accommodation platforms, online reviews serve as quality indicators, affecting their decisions to make a transaction. Sentiment analysis and text mining can be used to analyze these online reviews to identify various factors embedded in them that can influence how guests perceive lodging in the sharing economy. Peer-to-peer accommodation platforms can benefit from analyzing these aspects since they can utilize the results to streamline their operations and give customers better services. Current research on this domain has only identified a limited number of important factors, such as trust, quality, security, price, cleanliness, and indoor environmental quality. However, there can be many other factors that can affect the accommodation experience. These factors would require further attention. Therefore, in this study a dataset pertaining to the Airbnb platform was considered which contained a total of 401 964 review comments. Word cloud, frequency distribution, and topic modeling were used as data analysis techniques to identify various factors affecting accommodation experience. Results indicate that factors including location, safety, host-guest interaction, amenities, proximity to restaurants and transit options, and apartment uniqueness can be primarily taken into account to give superior services to their clients.Publication Embargo Group Formation and Communication of Multitasking Multi-Robots for Smart City Cleaning Process(IEEE, 2022-12-09) Dahanaka, D.M.S.J; Wijesooriya, A.I.E; Wickramasinghage, D.S.S; Bhaggya, G.V.C; Harshanath, S.M.B; Rajapaksha, U. U. SIn this research paper, we focus on how multitasking robots team up to clean a city. In particular, we consider how they build their team, how they position themselves in their positions, how they work with teams, how they face obstacles along the way, and how to move groups out of control in an emergency. We use a leader-follower strategy here, and we are also tasked with selecting a leader for each group. The leader finds the shortest route to avoid the obstacle by considering the obstacle details such as obstacle location, obstacle width, and destination. The leader decides the best way for the team to go. If the leader wants to change the group, it gives the message to the relevant member. In the event of meeting an obstacle, it changes its shape and moves. A Robot Operating System (ROS) framework was created to perform real-time experiments with ROS-capable mobile robotic TURTLEBOTs to evaluate this control strategy. Simulations performed on a mobile robot team demonstrate the effectiveness of the proposed approach.Publication Embargo DevFlair: A Framework to Automate the Pre-screening Process of Software Engineering Job Candidates(IEEE, 2022-12-09) Jayasekara, R.T.R; Kudarachchi, K.A.N.D; Kariyawasam, K.G.S.S.K; Rajapaksha, D; Jayasinghe, S.L; Thelijjagoda, SThe HR department of a technology company receives hundreds of job applications for each Software Engineering related vacancy. Evaluating a candidate by looking at the curriculum vitae may appear to be easy during the pre-screening process. However, an automated pre-screening process using Natural Language Processing and Machine Learning methodologies would help the recruiter to obtain a more accurate and deeper understanding of the candidate. In this paper we propose “DevFlair”, a framework to automate pre-screening Software Engineering job candidates. DevFlair uses data from social media, GitHub, and open-ended questionnaires to predict the Big-Five personality traits, analyze technical skill expertise, and analyze the experience in using industry-related online platforms. After analysis, the candidates are ranked according to their personality and technical skill levels. We conduct the personality prediction experiments using a social media posts dataset annotated with gold-standard Big-Five personality labels. We train FastText classification models and compare their accuracy against other state of the art classification models. The comparisons conclude that the FastText classification models substantially outperform the state of the art classification models when predicting Openness, Conscientiousness, and Agreeableness personality traits.Publication Embargo Solution to Measure Employee Productivity with Employee Emotion Detection(IEEE, 2022-12-09) De Silva, T.R.S.; Dayananda, K.Y.; Galagama Arachchi, R.C.; Amerasekara, M.K.S.B.; Silva, S; Gamage, NHealth and safety of workers has become a top priority in modern businesses. The reason is that it will have an impact on both individual and team output. In the last few decades, automatic facial expression analysis using machine learning has emerged as a promising and bustling field of study. In this study, the system primarily evaluates the efficiency of workers and, through the detection of their emotional states, determines their levels of motivation. The task completion rate of employees is measured by the system in the first component, and the system predicts the level of satisfaction that the employees will have. In place of linear regression, this component makes use of random forest regression, which boasts a higher degree of precision than its counterpart. The performance of workers on their tasks will be evaluated periodically, about once every fifteen minutes, and the results will be shown on a dashboard. The system will pick up on the emotions of the staff members throughout the second phase of the process. These characteristics will be used to assess the level of motivation inside the organization, with the end goal of increasing overall productivity. The accuracy of this emotion detection will also be checked periodically, namely once every fifteen minutes. The following part of the process monitors the use of the PC and calculates the level of productivity. It will be possible to get an increase in productivity if one monitors and keeps track of the application usage of each employee. The final components monitor the websites that employees visit and how they use the network. This component makes it easier to generate reports based on the utilization of the internet and the network, as well as information on performance and reports that summarize website traffic. When it is fully operational as an integrated system, most businesses will rely on this system as their primary driver of success.Publication Unknown Depression Detection System Using Real-Time and Social Media Data(IEEE, 2022-12-09) Jayasinghe, G.C.J.; Shamika, I.P.M.A.; Dissanayake, G.A.I.P; Ranaweera, R.M.I.A; Bandara, P.SThe main objective of this study is to measure the depression level of the participants. The guidance will be provided by the psychiatrist to understand the parameters. The end system has been implemented to measure it with a live session with pre-designed questionnaire set. During the session time, the behavior of the participant has been captured through audio and video method. The long-term depression level measurement will be analyzing the social media behavior of the participant within a month. The Convolution Neural Network (CNN) and Natural Language Processing (NLP) are using to analyze the video, audio and text data. To analyze the results; The Beck Depression Inventory (BDI II) scale will be utilized. The accuracy of the output results measured as high as it has been individually analyzed the subcomponents and then predict to a one result.Publication Embargo Human Behavior Analysis for Psychological Healthcare Sector (Project SERENITY)(IEEE, 2022-12-09) Dassanayake, D.M.H.; Wanigathunga, C.V.; Meeriyagalla, P.Y.; Yapa, K; Wickramarathne, K.A.P.P.; Rukgahakotuwa, LMental health is a key area of the healthcare sector. While taking care of the physical health of the human body, it is important to pay attention to mental health as well. This project is done to help people maintain their mental health. ‘ SERENITY’ is a web application designed not only for patients but also for doctors. This app works as a virtual assistant for a doctor, and this app helps doctors constantly monitor their patients’ behaviour, as well as SERENITY, which will be able to analyze the emotions of patients individually.Publication Embargo E-Learning Assistive System for Deaf and Mute Students(IEEE, 2022-12-09) Ranasinghe, P; Akash, K; Nanayakkara, L; Perera, H; Chandrasiri, S; Kumari, SE-learning has become a popular digital platform among both students and teachers. When using an e-learning system, deaf-mute students can get significant benefits. It allows students to better grasp their studies by providing additional details. The major problem that the deaf and mute community encounters in the e-learning environment is that they are no longer attempting to enroll in normal institutions, which do not provide many facilities for them due to a lack of resources, a lack of learning facilities, and some social disturbances. To achieve that problem this system will translate the lecturer’s voice into text, map words with pre-created sign language animations, generate subtitles for lecture videos, clearly identify the face position of the lecturer, detect difficult words, track the hand gestures, and practice sign language so that it will increase learning resources, facilities, usability and help teachers to execute their teaching process through this platform. Therefore, normal institutions can use this system as their learning management system. It will approach deaf and mute students to enroll in normal institutions and do their studies as typical students.Publication Embargo Sinhala Named Entity Recognition Model: Domain-Specific Classes in Sports(IEEE, 2022-12-09) .Wijesinghe, W.M.S.K; Tissera, MNamed Entity Recognition (NER) is one of the crucial and vital subtasks that must be solved in most Natural Language Processing (NLP) tasks. However, constructing a NER system for the Sinhala Language is challenging. Because it comes under the category of low-resource languages. Therefore, the proposed approach attempted designing a mechanism to identify specific named entities in the sports domain. Firstly, a domain-specific corpus was built using Sinhala sport e-News articles. Then a semi-automated, rule-based component named as “Class_Label_Suggester” was built to annotate pre-defined named entities. After auto annotation, the outcome was further validated manually with a little effort. Finally, it was trained using the annotated data. Linear Perceptron, Stochastic Gradient Descent (SGD), Multinomial Naive Bayes (MNB), and Passive Aggressive classifiers were used to train the NER model. Though, the above Machine Learning (ML) algorithms showed approximately 98% accuracy, the MNB model demonstrated highest accuracy for the identified class labels of which, 99.76% for ‘Ground’, 99.53% for ‘School’, 98.55% for ‘Tournament’, and 97.87% for ‘Other’ classes. Additionally, high precision values of the above classes were 81%, 72%, 62%, and 98% respectively. An accurately annotated Sinhala dataset and the trained Sinhala NER model are main contributions of the study.Publication Embargo ELIZA: Smart Monitoring and Reporting Toast Master System(IEEE, 2022-12-09) Nizer, F.S.A; Iksudha Bhargavi, R; Agalyah, P; Raveendran, M; Kuruppu, A; Rupasinghe, SPublic speaking is the most common form of fear, and everyone feels uneasy with it. Fear of speaking in public is commonly called “glossophobia,” where people are discouraged from speaking in front of people due to embarrassment and rejection. Public speaking anxiety (PSA) is one of the most universal subtypes of anxiety where people fear, lose their confidence, and become uncomfortable physically and mentally. But public speaking is considered important in the educational sector and workplaces, where people get higher opportunities. Therefore, clubs like Toastmasters help people overcome their fear of public speaking and improve their confidence. We are launching the idea of a Smart Monitoring and Reporting Toastmasters System for people to improve their public speaking so they do not need a supervisor or mentor to train them. This smart monitoring system recognizes the candidate through image processing and deep learning. Moreover, this will analyze some features from the candidates’ speeches, such as facial emotion recognition, speech recognition, hand and body gesture recognition, and the candidates’ attire and appearance separately. This system will identify their mistakes and flaws and provide overall feedback to the users on the speech provided by the candidate. By implementing this web application, users can train themselves without a supervisor, and they can improve themselves and gain the confidence to participate in a Toastmasters competition as perfect candidates.Publication Embargo LAWSUP - A Smart Platform to Assist Stakeholders of Business Law(IEEE, 2022-12-09) Sulakshi, U L H; Opatha, S D; De Silva, K S D; Sandeepa, M M A D N; Nawinna, D; Harasgama, H; Gamage, NCorporate law, sometimes known as business law, is the body of law that governs the rights, relationships, and behavior of persons, corporations, organizations, and businesses. Business Organizations, employees/laborers, and the public are involved in this area of the law accompanying lawyers, and legal advisors. Business organizations need legal advice. Employees face many difficulties and injustices at their workplaces. People who wish to start a new business, search for legal guidance. When one of these parties needs support, they must seek a lawyer, go to the lawyer, and get legal support. When delivering legal support to clients, lawyers are still going through a manual process. There are very few systems that have been implemented for the law domain so far, and those only search engine types of systems that are unable to support every stakeholder of this domain. There is no common platform for all these stakeholders to find solutions, connect with a good lawyer and get support. We have identified the main issues faced by business organizations, employees that need legal support, the general public, and lawyers, and developed a web solution by implementing Machine Learning, Classification Algorithms, Text mining, Natural Language Processing, and Web Crawlers.Publication Embargo Smart Caring System for Ornamental Fish(IEEE, 2022-12-09) Fernando, S; Jayaweera, N; Pitawala, S; Kaushalya, R; Ratnayake, P; Siriwardana, SOrnamental Fish Industry continues to be one of the fastest growing sectors worldwide. Healthy fish production at aquariums requires intensive care and ensures a stable and an optimum production environment inside the fish tanks, which is a challenging task. Unfortunately, due to the limitations in fish industry, productivity of well-developed, healthy fish has drastically depreciated. Limited skills and knowledge of aquarists have been a challenging task which has led to inaccurate predictions on certain factors such as quantification and length of estimation, amounts and types of fish food and servicing the filters at proper time intervals. Existing aquariums depend on the experience and availability of the aquarists, which can be a challenging process in real life. Developing a system to regulate these major concerns is a prominent solution. This research is done to propose an automated method, with the help of several fish aquariums and existing research papers, to encounter the mentioned major concerns which affects the aquarists and other stakeholders.Publication Embargo Guardian - Smart Assistant Tool for Visually Impaired People(IEEE, 2022-12-09) Amarasinghe, C.K.; Pinto, R.D.S.P.; Sudusinghe, K.N.At present, with the advancement of technology, various devices and solutions have been found to aid the visually impaired community (VI Community). Even with the countless technological breakthroughs, yet they face many problems performing the most basic functions in daily life. Identifying the objects, they use daily, identifying a person, whether it’s someone they know or not, and their emotions, and reading a text information displayed anywhere, without the assistance from another person are the basic issues we deal with and try to resolve using a tool consisting of a pair of spectacles with an inbuilt camera that is integrated with a mobile application. The inbuilt camera will capture the image of a text containing label, an object, or a person which will be then detected, analyzed, and recognized and will be converted to Speech using Google TTS engine and produced through the headphones giving the output to the user. Tesseract OCR, YOLO algorithm, and TensorFlow models have been used for each feature of the tool respectively. This tool will be very beneficial to a blind person as it mitigates them the frustration caused by being incapable of performing daily activities without any assistance from another person.Publication Embargo “Stonelia” – Prehistoric Stone Tool Identification Android App for Archaeological Researchers(IEEE, 2022-12-09) Perera, K. B. N.; Pathirana, M. K.; Wijayarathna, P. G. R. C.; Amarasena, N; Hewabathma, C. D.Prehistoric stone tools can be considered one of the oldest artifacts created by ancient humans. Lithic archeology’s study of stone tools provides important information about early humans’ technologies, agility, and mental and innovative abilities. A vital issue in lithic archeology is the identification and analysis of stone tools found at the excavation sites. Archeologists need to observe and analyze a stone tool under different aspects for a long time to verify whether it is a stone tool or a geofact, the techniques used to create it, and identify its rough relative date and functional value. This can be challenging for amateur scholars studying archeology since it requires a lot of experience and time to identify by a glance. As a solution, ‘Stonelia,’ a mobile-based android application, can be introduced to identify and analyze stone tools. The images captured through the mobile app are preprocessed using image processing. Using Convolutional Neural Network models identifies the stone artifact from a geofact, the mineral type, the rough relative date, techniques used to create the stone artifact, and its functional value. This mobile application provides prompt identification and analysis of stone artifacts within a short time and with higher accuracy.Publication Embargo Energy and Operations Optimization for Effective Greenhouse Management(IEEE, 2022-12-09) Prihan Nimsara, K. I.; Bodaragama, J.; Roshan Maduwantha, K. A.; Fernando, S. D.IoT technology-based process automation that can be applied to a greenhouse leads to making condition management and status monitoring more robust while leading to saving energy and resources. The proposed system which is based on IoT technology and MQTT protocol can set optimal growth conditions for plant and seed growth within the greenhouse. The sensor-based inputs are to be transformed into the processed values based on the defined logic and the standard benchmarks gathered from the local agricultural authorities. The key areas of condition monitoring to be done via temperature, humidity, soil moisture, and lighting can ultimately yield an increased harvest having supported both the plant and seeds-based implementations for multiple types of plants. One of the most important factors to consider is that the farmers can have energy savings through the proposed solution by controlling the actuators in an optimal manner and reducing manual intervention by a considerable amount. The excess usage of electricity by lights and cooling fan usage in the greenhouse can be controlled with real-time data tracking and better analytics. The use of water can be properly maintained for the plants by putting only the required amount will make the soil wet and spraying the required amount to air will make better humidity control. Thus, the real-time condition-based controlling of the actuators leads to making the greenhouse operations more optimal and better utilization of resources and energy which ultimately results in financial benefits for the greenhouse owner. Based on the evaluated power consumption of the greenhouse power usage before and after the system was installed, the newly introduced system can save energy by having optimal control of actuators by performing algorithmic calculations to meet only the required level of weather conditions. This is to be proven experimentally by implementing the proposed system for a defined period of time under the monitoring of energy usage.Publication Embargo Investigation of Routing Techniques to Develop a Model for Software-Defined Networks using Border Gateway Protocol(IEEE, 2022-12-09) Dayapala, B; Palanisamy, V; Suthaharan, SIn many cases, children between this age are using smartphones and other technology devices, to play games, watch cartoons, take photos and sometimes the chance is getting higher than we think that children access unnecessary contents due to lack of guidance and unawareness of parents. This interactive mobile application is used as an adaptive learning tool for the primary school students. Utilizing children’s comfort with technology allows for the development of their talents. In math skills development, some attractively designed gamified activities to solve basic math questions are given according to the skill level the child is currently in. The accuracy was much higher in the Convolutional Neural Network approach as it recorded a value of 0.9919. In environmental skills development component, the app will ask child to identify the surroundings according to a flow, starting from the house and towards the garden using object detection and the results were detected with a higher accuracy level around 0.9-0.99 after training the Machine Learning model. And in the language skills development component the child is given activities to develop pronunciation skills using audio processing and finally the verification of online achievements of a child by Non-Fungible Token technology, is fulfilled via the app.Publication Embargo Anomaly Detection in Microservice Systems Using Autoencoders(IEEE, 2022-12-09) de Silva, M; Daniel, S; Kumarapeli, M; Mahadura, S; Rupasinghe, L; Liyanapathirana, CThe adaptation of microservice architecture has increased massively during the last few years with the emergence of the cloud. Containers have become a common choice for microservices architecture instead of VMs (Virtual Machines) due to their portability and optimized resource usage characteristics. Along with the containers, container-orchestration platforms are also becoming an integral part of microservice-based systems, considering the flexibility and scalability offered by the container-orchestration media. With the virtualized implementation and the dynamic attribute of modern microservice architecture, it has been a cumbersome task to implement a proper observability mechanism to detect abnormal behaviour using conventional monitoring tools, which are most suitable for static infrastructures. We present a system that will collect required data with the understanding of the dynamic attribute of the system and identify anomalies with efficient data analysis methods.Publication Embargo Enhancing Conversational AI Model Performance and Explainability for Sinhala-English Bilingual Speakers(IEEE, 2022-12-09) Dissanayake, I; Hameed, S; Sakalasooriya, A; Jayasinghe, D; Abeywardhana, L; Wijendra, DNatural language processing has become essential to modern conversational tools and dialogue engines, including Chatbots. However, applying natural language processing to low-resource languages is challenging due to their lack of digital presence. Sinhala is the native language of approximately nineteen million people in Sri Lanka and is one of many low-resource languages. Moreover, the increase in using code-switching: alternating two or more languages within the same conversation, and code-mixing: the practice of representing words of a language using characters of another language, has become another major issue when processing natural languages. Apart from natural language processing, the explainability of opaque machine learning models utilized in chatbots has become another prominent concern. None of the existing modern chatbot development platforms supports explainability and relies on a performance score such as accuracy or f1-score. This paper proposes a no-code chatbot development platform with a series of built-in novel natural language processing, model evaluation, and explainability tools to tackle the problems of processing Sinhala-English code-switching and code-mixing natural language data and model evaluation in modern chatbot development platforms.Publication Embargo AI-Based Child Care Parental Control System(IEEE, 2022-12-09) Jayasekara, U; Maniyangama, H; Vithana, K; Weerasinghe, T; Wijekoon, J; Panchendrarajan, RDue to the prevalence of the COVID-19 epidemic around the globe, children were compelled to engage in remote learning through online platforms, hence mobile phone has become one of their predominant devices. Mobile device with Internet access offers a major outlet for education, entertainment, and social connection, but this combination can lead to several significant bad sequences such as online exploitation, harmful addictions, and other negative impacts of online social networking. To address harmful effects, parental controls are becoming more crucial, yet Sri Lankan parents are less aware of this. Consequently, this study proposes a parental control system to monitor their child’s activities. Android, Microsoft Azure, Java, Python, OpenCV, MySQL, and FastAPI are among the most prominent technologies utilized in the proposed application’s development. The suggested approach focuses primarily on the Sri Lankan context and aims to enhance parental digital literacy while safeguarding children from cyber threats. Yielded results showed the proposed mobile application for the identification of toxic words, drugs & alcohol content, game character images, and Instagram Sinhala comments severity as 94%, 95%, 97%, and 55% respectively in controlled experiments.Publication Embargo Smart UML - Assignment Management Tool for UML Diagrams(IEEE, 2022-12-09) Weerasinghe, D.N.H; Thiwanka, K.A.T; Jayasith, H.B.C; Onella Natalie, P.A.D; Rajapaksha, U.U.S; Karunasena, AThe great majority of manual processes have been automated in the current day. Nonetheless, there is no reliable approach for evaluating UML diagrams for plagiarism and correctness. Unified Modeling Language (UML) provides the basis for UML diagrams. UML Diagrams are typically used to graphically express a system’s stakeholders, activities, objects, and classes to get a deep knowledge of a particular system. UML diagrams are widely used by industry experts as well as beginners. Class diagrams and use case diagrams are the most commonly used diagrams in UML. Though most Computer science-related students get assessed in UML modelling, diagram plagiarism isn’t considered in most exams since it’s a tedious task to perform. The proposed system automates capturing details in UML diagram components and generating a model answer from the given question scenario to compare the diagrams and calculate the plagiarism percentage with the correction percentage.