2nd International Conference on Advancements in Computing [ICAC] 2020
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/1317
Browse
Publication Embargo AI - Driven Smart Bin for Waste Management(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Abeygunawardhana, A. G. D. T.; Shalinda, R. M. M. M.; Bandara, W. H. M. D.; W. D. S. Anesta, D.; Kasthurirathna; Abeysiri, L.With increasing urbanization, waste has become a major problem in the present world. Therefore, proper waste management is a must for a healthy and clean environment. Though government authorities in most countries provide various solutions for waste management, solid waste tends to make a significant impact on the environment as they do not decompose easily. This research focuses on AI (Artificial Intelligence)-driven smart waste bin that can classify the most widely available solid waste materials namely Metal, Glass, and Plastic. The smart waste bin performs the separation of waste using image processing and machine learning algorithms. The system also performs the continuous monitoring of the collected waste level by using ultrasonic sensors. A dedicated mobile application will generate the optimal routes for the available waste collectors to collect the filled bins. Moreover, with this smart bin, the challenge of recognizing each waste item is overcome by using visual data as the source. Therefore, the usage of expensive sensor devices and filtration techniques to determine the category is disregarded. The smart bin can recognize the category of solid waste, collect it to the specified container, and notify the garbage level in each container. So, it is a portable waste management system.Publication Embargo AI Approach In Monitoring The Physical And Psychological State Of Car Drivers And Remedial Action For Safe Driving(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Shanmugarajah, S.; Tharmaseelan, J.; Sivagnanam, L.Road Accidents and casualties incited by drowsiness are an overall important social and monetary issue. The connection between drowsiness and accidents is bolstered by logical confirmations that relate to small-scale sleep. This project has focused on Driver drowsiness detection by using ECG signal extraction. This work expects to extract and arrange the basic four types of sleep through Wavelet Transform and machine learning calculations. The report covers a short theoretical introduction about the medicinal topic, features the extraction, filtering techniques, and afterward trains the extracted information through machine learning software. After that is covered, it demonstrates the results with two types of machine learning algorithms (active or drowsiness status) with WEKA software. The main benefit of this system is it will send a notification to the driver's mobile every second when he goes to sleeping status. Nowadays artificial intelligence cars are available with sleep assistance, however, the devices used on these cars are very expensive. So, our approach is to develop a system to predict the driver's drowsiness to reduce accidents caused by sleepiness at a low cost. The sleep / awake status is determined by both the factors RR peak's distance and R's amplitude.Publication Embargo Algorithms for Automatic Identification and Analysis of Sri Lankan Anopheles Mosquito Species(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Palanisamy, V.; Thiruchenthooran, V.; Noble Surendran, S.; Ratnarajah, N.Microscopic digital image processing algorithms are presented here to automatically detect primary morphological features of Sri Lankan anopheline mosquitoes, as an essential step towards the development of automated identification and analysis of various species of anopheline mosquitoes. Mosquitoes that belong to genus Anopheles spread the causative pathogen of malaria. Perfect and speedy species identification is crucial in any surveillance and control strategies. Currently, morphological taxonomic keys are used to identify various species. Two or more primary morphological characteristics, such as a number of dark spots of wings and pale bands of legs, are used in each step of the hierarchical key. To achieve the automatic detection of the primary morphological features, image processing algorithms performed at three levels. At the pre-processing level, methods work with raw, possibly noisy pixel values, with noise reduction and smoothing. In the mid-level, algorithms are utilized pre-processing results for further means with background removing and spots/bands segmentation. At the final level, techniques try to extract the semantics of spots/bands and counting the spots/bands from the information provided. Thirty samples of anopheline mosquitoes' wings and legs microscopic images were analysed with satisfactory results.Publication Embargo Aspect Based Sentiment Analysis for Evaluating Movies and TV series Publisher: IEEE Cite This PDF(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Cooray, T.; Perera, G.; Chandrasena, D.; Alosius, J.; Kugathasan, A.Aspect-based sentiment analysis (ABSA) is used in different fields for analyzing customer reviews to project an overall customer opinion on certain products. With the expansion of the internet, people are provided with an inexpensive and time-saving method to express their opinion to a larger audience, while various industries are handed with the opportunity to gather free information from it to obtain market value. The implementation of machine learning methods for the evaluation of aspects related to movies and television series has not been commenced, and it could be a new development for the industry. This study focuses on conducting an ABSA on a movie or a television series based on genre, story as well as cast and crew aspects. The data collected from social media through web scraping is processed to produce adequate results to get a broad understanding on how the popularity of the movie or the television series related to above mentioned aspects. Then, each aspect is further analyzed to gather precise information belonging to each aspect. The accuracy of the results of the proposed system has been achieved over 79%. The results proved that the solution is highly successful than the former works with high business value.Publication Embargo Assist: Rendering, Pipeline Management, and Pipeline Tracking Software Publisher: IEEE Cite This PDF(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Salgado, M.V.I; Hettiarachchi, H.A.D.D.; Munasinghe, T.U.; Fernando, K.A.U.; Gamage, I.; Thilakarathna, T.; Cooray, N.C.Video production is one of the most dominant industries in the 21st century, and research into the automation of tasks associated with it has drastically increased. The production of videos take place in three stages: pre-production, production, and post-production. These three stages consist of script writing, scheduling, logistics, and other administration work. There are commercial products to automate these individual tasks. Incorporating all these software into video production can be expensive and difficult to manage. This study proposes the “Assist” software to handle all processes in video production. It has resulted in a product that covers the three main stages featuring scripts, storyboards, inventory management, production progress tracking and management, and rendering. The mentioned features were designed and developed using decision tree algorithm, PyQt5, general decimation algorithm, mesh simplification algorithm, and multi-variable regression.Publication Embargo Behavior Segmentation based Micro-Segmentation Approach for Health Insurance Industry(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Nandapala, E.Y.L.; Jayasena, K.P.N.; Rathnayaka, R.M.K.T.To manage the company’s future growth, the relationship between companies and customers is important. This can be referred to as Customer Relationship Management (CRM). By applying the micro-segmentation process companies can succeed in this CRM process. Micro-segmentation is a breakdown into micro-segments of the entire data collection. The user can easily be deeply defined with this segmentation process. Demographic segmentation is a breakdown of the dataset based on the consumers’ age, gender, etc. Behavior segmentation is diving the whole dataset based on customers’ behaviors. RFM analysis is a behavioral segmentation process based on consumer’s behaviors. There is no exact way to precisely conduct micro-segmentation. Thus, this study proposed a new micro-segmentation process. That is applying demographic segmentation with the support of the RFM analysis. This method can easily determine the customers’ behaviors accurately and deeply. Insurance companies offer different types of insurance and health insurance is the most critical insurance type for humans. By applying the proposed method in this research, health insurance companies can determine the policyholder’s behaviors, claiming patterns, claiming chargers, and other information precisely. Furthermore, health insurance providers can effectively manage their claims using this knowledge.Publication Embargo Blockchain based Patients' detail management System(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Abeywardena, K.Y.; Attanayaka, B.; Periyasamy, K.; Gunarathna, S.; Prabhathi, U.; Kudagoda, S.In the data technology revolution, electronic medical records are a standard way to store patients' information in hospitals. Although some hospital systems using server-based patient detail management systems, they need a large amount of storage to store all the patients' medical reports, therefore affecting the scalability. At the same time, they are facing several difficulties, such as interoperability concerns, security and privacy issues, cyber-attacks to the centralized storage and maintaining adhering to medical policies. Proposed Flexi Medi is a private blockchain based patient detail management system which is expected to address the above problems. Solution proposes a distributed secure ledger to permits efficient system access and systems retrieval, which is secure and immutable. The improved consensus mechanism achieves the consensus of the data without large energy utilization and network congestion. Moreover, Flexi Medi achieves high data security principles based on a combination of hybrid access control mechanism, public key cryptography, and a secure live health condition monitoring mechanism. The proposed solution results in successfully deployed smart contracts according to the roles of the system, real time patient health monitoring with more scalable and access controlled system. The overall objective of this solution is to bring the entire medical industry into a common platform using a decentralized approach to store, share medical details while eliminating the need to maintain printed medical records.Publication Embargo Candidate Selection for the Interview using GitHub Profile and User Analysis for the Position of Software Engineer(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Gajanayake, R.G.U.S.; Hiras, M.H.M.; Gunathunga, P.I.N.; Supun, E.G.J.; Karunasenna, A.; Bandara, P.Selecting the most suitable candidates for interviews is an important process for organizations that can affect their overall work performance. Typically, recruiters check Curriculum Vitae (CV), shortlist them and call candidates for interviews which have been the way of recruiting new employees for a long time. To minimize the time spent on the above process, pre-screening mechanisms are nowadays implemented by organizations. However, those mechanisms need sufficient information to evaluate the candidate. For example, in case of a software engineer, the recruiters are interested on the programming ability, academic perfo rmance as well as personality traits of potential candidates. In this research, a pre-screening solution is proposed to screen the applicants for the post of Software Engineer where candidates are screen based on an initial call transcript, GitHub profile, LinkedIn profile , CV, Academic transcript and, Recommendation letters. This approach extracts textual features of different dimensions based on Natural Language Processing to identify the Big Five personality traits, CV and GitHub insights, candidate’s skills, background, and capabilities from Recommendation letters as well as programming skills and knowledge from Academic transcript and Linked Profile. The results obtained from the different areas are presented an d shown that the selected supervised machine learning algorithms and techniques can be used to evaluate the best possible candidates.Publication Embargo Candidate Selection for the Interview using GitHub Profile and User Analysis for the Position of Software Engineer(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Gajanayake, R.G.U.S.; Hiras, M.H.M.; Gunathunga, P.I.N.; Supun, E.G.J.; Karunasenna, A.; Bandara, P.Selecting the most suitable candidates for interviews is an important process for organizations that can affect their overall work performance. Typically, recruiters check Curriculum Vitae (CV), shortlist them and call candidates for interviews which have been the way of recruiting new employees for a long time. To minimize the time spent on the above process, pre-screening mechanisms are nowadays implemented by organizations. However, those mechanisms need sufficient information to evaluate the candidate. For example, in case of a software engineer, the recruiters are interested on the programming ability, academic perfo rmance as well as personality traits of potential candidates. In this research, a pre-screening solution is proposed to screen the applicants for the post of Software Engineer where candidates are screen based on an initial call transcript, GitHub profile, LinkedIn profile , CV, Academic transcript and, Recommendation letters. This approach extracts textual features of different dimensions based on Natural Language Processing to identify the Big Five personality traits, CV and GitHub insights, candidate’s skills, background, and capabilities from Recommendation letters as well as programming skills and knowledge from Academic transcript and Linked Profile. The results obtained from the different areas are presented an d shown that the selected supervised machine learning algorithms and techniques can be used to evaluate the best possible candidates.Publication Embargo Character Modifier Combinations Recognition in Sinhala Handwriting(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Silva, C.M.; Jayasundere, N. D.Sinhala script is categorized as a segmental writing system and therefore consonant–vowel sequences are identified as a unit. Depending on the vowel and the consonant combination, the diacritic can attach above, below, following or preceding the consonant. The identification of characters with the modifier symbols is important in recognizing generally written Sinhala text and has not been addressed sufficiently in the existing research. Sinhala handwriting recognition is much difficult compared to the other popular languages due to the complexity of the shapes of the characters and the presence of the modifier symbols. This paper discusses on a projection profile, distance profile, partial distance profile and contour based approach to identify characters along with character modifiers in Sinhala script. The proposed method has given an average recognition rate of 75% for 283 character modifier combinations. The proposed solution can be used to identify Sinhala handwritten text with a proper segmentation mechanism.Publication Embargo ChildPath: Diagnose depression in pre-schoolers based on daily activ(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Kirthika, L.; Abeykoon, J.To determine depression in pre-schoolers and validation of identifying depression based on daily activities. A comprehensive literature search, interviews with accredited mental health practitioners and a survey was conducted to validate the background aspects and existing diagnosis theories to map out based on daily activities. The results of the evaluation suggest a gap around diagnosis of depression in pre-schoolers due to lack of awareness and its distinctive nature to adult depression. This establishes a need for depression status calculation mechanism based on analysis of daily activities using machine learning to examine behaviour and speech patterns. Further, rule-based machine learning, will be implemented to offer personalized treatment plans if diagnosed with a status of depression.Publication Embargo Computational Model for Rating Mobile Applications based on Feature Extraction(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Gunaratnam, I.; Wickramarachchi, D.N.Google Play Store and App Store allow users to share their opinions and helps to measure users satisfaction level about the app through user comments. However, it's highly time-consuming to process all reviews manually. The usefulness of star ratings is limited for development teams since a rating represents an average of both positive and negative evaluations. Therefore, an automated solution is needed to systematically analyze reviews and other textual forms of data. The main objective of this research is to build a platform that rate apps by feature extraction and sentiment analysis to calculate the functionality index of apps based on metrics obtained by surveying 204 mobile phone users. The 5 topmost metrics obtained from them among the 16 metrics obtained from the literature review are usability, price, and frequency of updates, ad-freeness and battery consuming level. This research focuses on selected apps in music and audio category. To perform app rating indexes calculation of the overall app's reviews; data extraction, data cleaning, POS tagging, feature extraction, feature/feature values pairing, weighted feature rating, overall apps' rating and feature-wise app rating is done on textual data. The accuracy of the created model is measured by the level of satisfaction from users.Publication Embargo A Deep Learning Approach to Outbreak related Tweet Detection(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Jayawardhana, B. A. S. S. B.; Rajapakse, R. A. C. P.Due to the popularity of social media around the world, people use to report and discuss real-world events, personal health complications, and disaster situations through these platforms. These social media data streams can be used to track and detect different types of outbreaks. A mechanism is needed to identify outbreak-related tweets to predict the outbreak in advance. In this paper, we propose a deep learning model that can detect tweets related to different outbreaks Epidemics, Public Disorders, and Disasters. GloVe (Global Vectors for Word Representation) embeddings are used as the feature extraction technique as it can capture the semantic meanings of the tweets. Long Short-term Memory (LSTM) which is a specialized Recurrent Neural Network architecture is used as the classification algorithm. In the process, first, outbreak-related tweets were manually collected and curated. Pretrained GloVe word embeddings of 100 dimensions were then used to represent the words of the tweets. As the next step, a Deep Learning Model was trained by using LSTM technique on the curated dataset. Finally, the performance of the model was evaluated using a different dataset. With the results, it can be concluded that the proposed deep learning model is an accurate approach for outbreak-related tweet detection.Publication Embargo Deepfake Audio Detection: A Deep Learning Based Solution for Group Conversations(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Wijethunga, R.L.M.A.P.C.; Matheesha, D.M.K.; Al Noman, A.; De Silva, K.H.V.T.A.; Tissera, M.; Rupasinghe, L.The recent advancements in deep learning and other related technologies have led to improvements in various areas such as computer vision, bio-informatics, and speech recognition etc. This research mainly focuses on a problem with synthetic speech and speaker diarization. The developments in audio have resulted in deep learning models capable of replicating naturalsounding voice also known as text-to-speech (TTS) systems. This technology could be manipulated for malicious purposes such as deepfakes, impersonation, or spoofing attacks. We propose a system that has the capability of distinguishing between real and synthetic speech in group conversations.We built Deep Neural Network models and integrated them into a single solution using different datasets, including but not limited to Urban- Sound8K (5.6GB), Conversational (12.2GB), AMI-Corpus (5GB), and FakeOrReal (4GB). Our proposed approach consists of four main components. The speech-denoising component cleans and preprocesses the audio using Multilayer-Perceptron and Convolutional Neural Network architectures, with 93% and 94% accuracies accordingly. The speaker diarization was implemented using two different approaches, Natural Language Processing for text conversion with 93% accuracy and Recurrent Neural Network model for speaker labeling with 80% accuracy and 0.52 Diarization-Error-Rate. The final component distinguishes between real and fake audio using a CNN architecture with 94% accuracy. With these findings, this research will contribute immensely to the domain of speech analysis.Publication Embargo DenGue CarB: Mosquito Identification and Classification using Machine Learning(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Mohommed, M.; Rajakaruna, P.; Kehelpannala, N.; Perera, A.; Abeysiri, L.This research paper discusses a web-based application that assists Public Health Officers in the dengue identification process. The mosquito classification is done using image processing and machine learning techniques. The training models are developed using Convolutional Neural Networks Algorithm, Support Vector Machine Algorithm, and K-Nearest Neighbors Algorithm to validate the results to determine the most accurate and suitable algorithm. this paper discusses the previous related research work on its significance and drawbacks while highlighting design, methods, and implementation in the solution. We conclude that the CNN algorithm provides the highest accuracy among the machine learning techniques used.Publication Embargo DNN Based Currency Recognition System for Visually Impaired in Sinhala(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Gamage, C.Y.; Bogahawatte, J.R.M.; Prasadika, U.K.T.; Sumathipala, S.Recently researches have been conducted in the domain of currency recognition. The task of recognizing the currency notes has become challenging due to the distortion of the notes over time. Currency recognition systems in Sinhala for visually impaired people are rarely developed. To address this problem a research has been done and a relevant application has been implemented comprising three modules as Speech Recognition module, Currency Recognition module and Text to Speech Module. The major challenge in all three modules is to achieve a better accuracy using deep learning concepts. TensorFlow platform and Keras library were used to build the speech recognition neural network model for Sinhala spoken words. Deep learning neural networks were utilized for the development of currency recognition module and text to speech module.Publication Embargo Domain Specific Conversational Intelligence: Voice Based E-Channeling System(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Weerathunga, W.A.H.; Lokugamage, G.N.; Hariharan, V.; Yahampath, A.D.N.H.; Kasthurirathna, D.In this research the application of Automatic Speech Recognition, Natural Language Understanding, Neural Networks and Text To Speech Conversion is investigated to create a domain specific end to end voice based E-Channeling system. The novel idea in this research can be extended to any other domain(e.g.: Taxi Application) and build a conversational intelligence system. This system enables the user to avoid the shortcomings in the traditional doctor appointment channeling procedures. The system also have the ability to predict the doctor specialization according to the symptoms of the patient and can give emergency health tips by using the powerful Neural Network module. Domain-specific speech recognition model is created according to Sri Lankan accents and handles the context-specific to this domain(94% accuracy). Extracting the entities, handling e-channeling functions and selecting the most suitable API is done by the RASA backend. Neural Network will give the first aid and doctor specialization recommendations according to user input with a validation accuracy of 90%. Speech synthesis model will output the response in user preferred language(Sinhala, English or Tamil).Publication Embargo EasyTalk: A Translator for Sri Lankan Sign Language using Machine Learning and Artificial Intelligence(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Kumar, D.M.; Bavanraj, K.; Thavananthan, S.; Bastiansz, G.M.A.S.; Harshanath, S.M.B.; Alosious, J.Sign language is used by the hearing-impaired and inarticulate community to communicate with each other. But not all Sri Lankans are aware of the sign language or verbal languages and a translation is required. The Sri Lankan Sign Language is tightly bound to the hearing-impaired and inarticulate. The paper presents EasyTalk, a sign language translator which can translate Sri Lankan Sign Language into text and audio formats as well as translate verbal language into Sri Lankan Sign Language which would benefit them to express their ideas. This is handled in four separate components. The first component, Hand Gesture Detector captures hand signs using pre-trained models. Image Classifier component classifies and translates the detected hand signs. The Text and Voice Generator component produces a text or an audio formatted output for identified hand signs. Finally, Text to Sign Converter works on converting an entered English text back into the sign language based animated images. By using these techniques, EasyTalk can detect, translate and produce relevant outputs with superior accuracy. This can result in effective and efficient communication between the community with differently-abled people and the community with normal people.Publication Embargo Enhanced Symmetric Cryptography for IoT using Novel Random Secret Key Approach(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Sittampalam, G.; Ratnarajah, N.The deployment of IoT devices in different domains enables technical innovations and value-added services to users but also creates multiple requirements in terms of effective communication and security. IoT devices are constrained by less computing resources and limited battery power. Generally, the TLS/SSL protocol is used to provide communication security on IoT and the protocol utilizes important encryption algorithms like RSA, Elliptic Curve Cryptography, and AES. However, these conventional encryption algorithms are computationally and economically expensive to implement in IoT devices. Lightweight Cryptography (LWC) algorithms were introduced recently for IoT and the aim of the algorithms is to provide the same level security with a minimal amount of computing resources and power. This paper proposes a novel Random Secret Key (RSK) technique to provide an additional security layer for symmetric LWC algorithms for IoT applications. In RSK, IoT devices do not transmit keys over the network; they share a random matrix, calculate their own RSK, encrypt, and transmit the cipher text. When a random matrix lifetime expires new matrix published and RSK resets. Regular change in the RSK makes the IoT networks resistant to brute-force/dictionary attacks. The RSK added one more simple and effective secure layer to strengthen the security of the original secret key and is successfully implemented in a smart greenhouse environment. The outcomes of the experiments prove that the RSK provides enhanced and efficient protection for symmetric LWC algorithms in any IoT systems, consume a minimum amount of resources and more resistant to key-based attacks.Publication Embargo An Enhanced Virtual Fitting Room using Deep Neural Networks(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Ileperuma, I.C.S.; Gunathilake, H.M.Y.V.; Dilshan, K.P.A.P.; Nishali, S.A.D.S.; Gamage, A.I.; Priyadarshana, Y.H.P.P.As the customer's experience in present fit-on rooms is considered as an essential part of the textile industry, these fit-on rooms play a huge role in the textile shops. It is quite an arduous method and generates problems like long queues, having to change clothes individually, privacy problems and wasting time. The proposed convolutional neural network-based Virtual Fit-on Room helps to prevent the above mentioned problems. This product contains a TV screen, two web cameras, and a PC. It captures the customer's body by using two web cameras and displays the customer's dressed body. The combination of CNN in Deep learning and AR processes the body detection and generates the customer's dressed object. The application uses the stereo vision concept to get body measurements. The system detects customer age, gender, face type, and skin tones which are used to recommend cloth styles to customers. Another requirement of this system is customizing styles according to the customer requirements and suggests different styles of clothes. The system achieved 99% accuracy when suggesting different styles using FFNN. Customers can choose clothes for another person who does not physically appear with the customer in the textile shop. The expected output delivers the most realistic dressed object to the customer which allows the efficient customizations for the textile products according to customer requirements. This product can highly influence the textile and fashion industry. Therefore, this product is suitable to compete with other applications in the industry.