2020
Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/2219
Browse
Publication Embargo Comparative analysis of the application of Deep Learning techniques for Forex Rate prediction(2019 1st International Conference on Advancements in Computing (ICAC), SLIIT, 2019-12-05) Aryal, S.; Nadarajah, D.; Kasthurirathna, D.; Rupasinghe, L.; Jayawardena, C.Forecasting the financial time series is an extensive field of study. Even though the econometric models, traditional machine learning models, artificial neural networks and deep learning models have been used to predict the financial time series, deep learning models have been recently employed to do predictions of financial time series. In this paper, three different deep learning models called Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN) and Temporal Convolution Network (TCN) have been used to predict the United States Dollar (USD) to Sri Lankan Rupees (LKR) exchange rate and compared the accuracy of the models. The results indicate the superiority of CNN model over other models. We conclude that CNN based models perform best in financial time series prediction.Publication Embargo Aerodynamic modeling of simplified wind turbine rotors targeting small-scale applications in Sri Lanka(Elsevier, 2020-09-11) Sugathapala, T. M; Boteju, S; Withanage, P. B; Wijewardane, SA design and optimization procedure of simplified wind turbine rotors for small-scale applications is presented. The need for this research has arisen from the recent national initiative of the government of Sri Lanka titled ‘Battle for Wind Energy’ in promoting small scale grid connected wind plants for electricity customers under Net Metering scheme. The main objective of this research is to assist local developers to design optimum rotors for given electrical generators (as determined by customer requirements), suitable for wind characteristics at specific locations. Another objective is to enhance local manufacturing capabilities by providing a design option of a simplified rotor blade geometry. A study on the correlation between population density of electricity customers and wind energy potentials was carried out to categorize the demand centres based on wind energy potentials in proposing series of small-scale wind turbine designs. A unique and improved rotor design procedure is presented which attempts to match the point of maximum performance of a rotor (design tip speed ratio) with the design wind speed of a given location by considering generator performance. The new design procedure showed successful convergence on a unique blade diameter for each rotor configuration that allowed the design tip speed ratio to match the design wind speed. The performance evaluation of rotor designs showed that high solidity rotors work better on the low wind potential region while low solidity rotors dominate medium and high wind potential regions. The performance reductions of simplified rotor designs are not significant and therefore would be an effective way to enhance value addition through local manufacture.Publication Open Access Comparative Analysis of Deep Learning Models for Multi-Step Prediction of Financial Time Series(researchgate.net, 2020-10-21) Aryal, S; Nadarajah, D; Rupasinghe, P.L; Jayawardena, C; Kasthurirathna, DFinancial time series prediction has been a key topic of interest among researchers considering the complexity of the domain and also due to its significant impact on a wide range of applications. In contrast to one-step ahead prediction, multi-step forecasting is more desirable in the industry but the task is more challenging. In recent days, advancement in deep learning has shown impressive accomplishments across various tasks including sequence learning and time series forecasting. Although most previous studies are focused on applications of deep learning models for single-step ahead prediction, multi-step financial time series forecasting has not been explored exhaustively. This paper aims at extensively evaluating the performance of various state-of-the-art deep learning models for multiple multi-steps ahead prediction horizons on real-world stock and forex markets dataset. Specifically, we focus on Long-Short Term Memory (LSTM) network and its variations, Encoder-Decoder based sequence to sequence models, Temporal Convolution Network (TCN), hybrid Exponential SmoothingRecurrent Neural Networks (ES-RNN) and Neural Basis Expansion Analysis for interpretable Time Series forecasting (N-BEATS). Experimental results show that the latest deep learning models such as NBEATS, ES-LSTM and TCN produced better results for all stock market related datasets by obtaining around 50% less Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) scores for each prediction horizon as compared to other models. However, the conventional LSTM-based models still prove to be dominant in the forex domain by comparatively achieving around 2% less error values.Publication Embargo Arogya -An Intelligent Ayurvedic Herb Management Platform(IEEE, 2020-11-04) Pathiranage, N; Nilfa, N; Nithmali, M; Kumari, N; Weerasinghe, L; Weerathunga, IAyurvedic means a science of life and well-being with its unique approaches to social and spiritual life. Especially in Sri Lanka we have our own set of rare Ayurvedic herbs which have been utilized by generations as medicinal treatments for a variety of diseases. Absence of specialists in this area makes proper identification as well as classification of valuable herbal plants a tedious task, which is essential for better treatment. Hence, a fully automated system for herb detection and classification, information visualization regarding them is highly desirable. There are existing applications which can identify plants with low prediction accuracies, as well as to give information regarding them. However, these applications are based on foreign plant data sets that do not include valuable herbs and shrubs with medicinal qualities. Hence this research proposes an application unique to medicinal plants, which can perform all these functionalities in both online and offline approach. Here, a new Ayurvedic plant dataset prepared from scratch, and preliminary results for classification of 5 types of herbs, compared with several deep Convolutional Neural Network (CNN) models based on transfer learning are presented. Experimental results indicate Marker-based Watershed algorithm as the best object detection algorithm in a complex background, VGG-16 as the best deep CNN classification model which reached a promising testing accuracy of 99.53%, and Seq2Seq LSTM model as the best deep learning model with optimum accuracy in abstractive information summarization.Publication Embargo VirtualPT: Virtual Reality based Home Care Physiotherapy Rehabilitation for Elderly(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Heiyanthuduwa, T.A.; Amarapala, K.W.N.U.; Gunathilaka, K.D.V.B.; Ravindu, K.S.; Wickramarathne, J.; Kasthurirathna, D.This paper describes the development of Personal computer based Virtual Reality home-care Physiotherapy system aimed for rehabilitating full body function in elders. VirtualPT is a true virtual reality platform where the environment is completely replaced by a virtual reality platform based on the mental condition of the person at the time. While doing the home-based prescribed physiotherapy exercises, the key health metrics are continuously monitored and tracked by combining the immersive Virtual Reality with the wearable VirtualPT Sensor kit. Virtual Reality combined with 3D motion capture lets real time movements to be accurately translated onto the virtual reality avatar that can be viewed in a virtual environment to assist physiotherapist to add exercises to the system easily. This ultimate virtual reality Physiotherapy assistant avatar is used to provide guidance to elders at home, to demonstrate and assist elders in adhering to the prescribed exercises. As a significant aspect of social interactions, mirroring of movements has been added to focus on whether the elder is able to accurately follow the movements of avatar. Furthermore, the insightful dashboard offers the elders and physiotherapists an interactive platform through virtual reality capabilities. VirtualPT physiotherapy system is cost effective and makes recovery and more convenient to elders at home while the participatory and immersive nature of Virtual Reality offers a unique realistic quality that is not generally existing in clinical-based physiotherapy. When looking at the broader concept of VirtualPT; continuity of care, integration of services, quality of life and access are equally important criteria which add more 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 Smart Intelligent Troubleshooter to Solve Windows Operating System Specific Issues(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Rajapakshe, D.I.K.; Shamil, M.P.P.; Paththinisekara, P.M.C.P.; Liyanage, S.K.; Samaratunge Arachchillage, U.S.S.; Kuruppu, A.While working on computers, people frequently confront with various kinds of problems, those beyond their extensive expertise. Microsoft Windows is the widely used Operating System running on numerous personal computers and the reason which gives more irritating problems that require to be addressed. Currently, troubleshooting is considered as a costly and time-consuming approach. The SAITA is an Artificial Intelligent Troubleshooting Agent that utilizes natural language generation, machine learning, and dependency resolving and ontology-based methodologies for solving most common Windows-specific issues within a short period of time than the traditional approach. The assistant learns from the accessible data and accomplishes the task for users as performed by human experts. The main objective of this exploration venture is to distinguish the constraints of existing troubleshooting software and create an AI troubleshooting assistant to provide solutions to fix the identified user issues. The use of this assistant would be economical as an IT help desk alternative in the industry. SAITA is developed to serve as a representative troubleshooter for fundamental user issues, service issues, application issues, and perform environment setup by analyzing software. This system will be able to solve the common Windows user’s issues as same as a human with less time.Publication Embargo IoT Based Sign Language Recognition System(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Punsara, K.K.T.; Premachandra, H.H.R.C.; Chanaka, A.W.A.D.; Wijayawickrama, R.V.; Abhayasinghe, N.; De Silva, R.Sign language is the key communication medium, which deaf and mute people use in their day - to-day life. Talking to disabled people will cause a difficult situation since a non-mute person cannot understand their hand gestures and in many instances, mute people are hearing impaired. Same as Sinhala, Tamil, English, or any other language, sign language also tend to have differences according to the region. This paper is an attempt to assist deaf and mute people to develop an effective communication mechanism with non-mute people. The end product of this project is a combination of a mobile application that can translate the sign language into digital voice and loT-enabled, light-weighted wearable glove, which capable of recognizing twenty-six English alphabet, digits, and words. Better user experience provides with voice-to-text feature in mobile application to reduce the communication gap within mute and non-mute communities. Research findings and results from the current system visualize the output of the product can be optimized up to 25 % -35 % with an enhanced pattern recognition mechanism.Publication Embargo SURAKSHA e-Caretaker: Elders Falling Detection and Alerting System using Machine Learning(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Mendis, L.; Hathurusinghe, S.; Epa, H.; Edirisinghe, T.; Wickramarathne, J.; Rupasinghe, S.People become unable to perform tasks that were done at the younger ages as they were when the ages pass with time. Falls play a major issue in the lives of elderly people as the physical and mental quality of life is dependable on the effects of falls. This research presents an e-Caretaker SURAKSHA which is an elder falling detection and alerting system based on Machine Learning concepts. Researchers that have been done in this area have produced different solutions to detect only the falls but not to automatically detect and notify them to the caretakers. This solution serves as a smart wearable device that is capable of automatically monitoring real-time postures, detecting sudden falls, possible arrhythmia conditions of the heart of the fallen person, and daily route deviations along with the fallen location which is finally notified to the caretakers through a mobile application. According to the performed studies, python model development was used to implement the system through Machine Learning concepts by referring to the Markov model, Prophet model, and Naïve Bayes algorithms. This solution provides the results of this research with an accuracy of around 89.9% leading to a successful product in the domain. Keywords—Publication Embargo Evaluating Teaching Content and Assessments Based on Learning Outcomes(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Pallegama, P.M.O.N.; Kumari, K.A.M.R.; Dissanayaka, D.M.D.P.M.; Ravihansi, A.V.Y.; Karunasenna, A.; Samarakoon, U.A modularized syllabus content assigned to different units of a subject proves very useful to both teachings as well as the student community. In each module, learning outcomes are defined. In each learning outcome, lesson learning outcomes are defined. When the Teaching Content (Lecture content), Learning activities (Labs sheets and Tutorials), Final Question Papers are being made the subject learning outcome should be considered and it should be made within the subject learning outcomes. Then the teaching and learning process will be done properly. Nowadays Revised Bloom's Taxonomy standard is used to structure the Teaching Content, Learning Activities, and Final Question paper of a course in the best way. Currently, there is no proper solution to corporate above areas according to the Revised Bloom's Taxonomy. This paper discusses an automated system that provides the features to verify the module and lesson learning outcomes and their levels according to Revised Bloom's taxonomy and to verify that the teaching content and learning activities are within the learning outcomes. Beyond that, this system uses various technologies and algorithms to improve the accuracy and efficiency of this research. This automated system is able to achieve to the final outcome with the best accuracy and efficiency than the manual process.Publication Embargo Generating 2.5D Motion Graphics from 2D Designs(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Ranawake, I.; Guruge, S.; Ahamed, A.; Kasthurirathne, D.As of the year 2020 , the video production industry is worth 31 billion dollars in the United States alone, with more than 6000 businesses and 57000 employees, and keeps growing. The global computer animation market size is anticipated to reach USD 28.30 billion by 2025, according to Grand View Research, Inc. Such significant growth demands the tech industry to introduce better tools for making animations. In this paper, we propose our contribution, specifically for User Interface designers in the field of motion graphics. UI/UX design is found to be one of the top 5 most trending opportunities for motion designers, and our proposed system allows them to generate a 2.5D animation based on a 2D Futuristic User Interface (FUI) design. The ultimate goal is to reduce the production time of FUI animations and minimize the cost of responding to the client’s changes. Obtaining clients’ feedback directly on animations rather than still images would improve the client’s involvement in production, resulting in greater confidence and loyalty. We implement several image processing techniques such as thinning and pixel clustering for pre-processing the 2D designs to segment the given design into an array of atomic shapes. Since the thinned shapes ensure that any pixel in the design does not have adjacent pixels which are also adjacent to each other, it is possible to utilize mathematical means to approximate the shapes. Our system converts a given 2D design to a collection of animated lines and arcs distributed in 3D space that eventually can be exported to the industry-standard tool, Adobe After Effects.Publication Embargo A Story of Two Surveys: for the Advancement of Sinhalese Mobile Text Entry Research(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Reyal, S.; Piyawardana, V.; Kaveendri, D.This paper presents two surveys: a literature survey on the current progress on Sinhalese mobile text entry research and a user survey on how Sri Lankans experience Sinhalese mobile text entry. The first survey concludes that Sinhalese mobile text entry is limited in scope and size compared to western text entry research. The second survey attempts to bridge this gap by providing deep insight into aspects in Sinhalese mobile text entry such as language switching, using English within Sinhalese e.g. mixed-mode and Singlish, and the popularity of various input modalities, keyboard vendors, and keyboard layouts. This is also the first research publication that unveils the current state-of-the-art in Sinhalese mobile text entry, along with user-preferences such as using autocorrect, glide-typing, and speech. Results from this survey deepens our understanding of the Sinhalese mobile text entry domain resulting in a stronger empirical footing and more innovative Sinhalese mobile text entry solutions.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.Publication Embargo Secured, Intelligent Blood and Organ Donation Management System - “LifeShare”(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Wijayathilaka, P.L.; Pahala Gamage, P.H.; De Silva, K.H.B.; Athukorala, A.P.P.S.; Kahandawaarachchi, K.A.D.C.P.; Pulasinghe, K.N.The scarcity and exigency for blood and organs has created many discrepancies in current approaches. These have created the criteria for malpractices such as organ trafficking and black market selling. This research presents a solution with a secured-smart blood and organ donation web developed system, allowing both patients and healthcare providers to access information about the blood and organ processing records. The database would be managed using the Blockchain technology which could be only accessed by authorized users. Finally, tracking all registered donors, the proposed system generates a smart identity developed by Ethereum Smart Contract (ESC). System predicts blood demand for the future ten years using Linear Regression Model with 0.998 of high R-squared accuracy value. This reduces shortages and wastage of blood. Also, using global positioning system and K-Nearest Neighbors Machine Learning algorithm, the system finds the best matches among donors and seekers according to the nearest location. Further, the system will automatically send questionnaires for registered users to identify and evaluate their awareness and issues about organ donation. Overall, this study aims for a secured and transparent web application. Thus, it facilitates an innovative and a productive blood donation and organ transplantation process in Sri Lankan healthcare sector.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 Smart Exam Evaluator for Object-Oriented Programming Modules(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Wickramasinghe, M.L.; Wijethunga, H.P.; Yapa, S.R.; Vishwajith, D.M.D.; Samaratunge Arachchillage, U.S.S.; Amarasena, N.Worldwide educators considered that, automate the evaluation of programming language-based exams is a more challenging task due to its complexity and the diversity of solutions implemented by students. This research investigates and provides insight into the applicability and development of a java based online exam evaluator as a solution to traditional onerous manual exam assessment methodology. The proposed system allows students to take online exams in Java for an implemented source code in a practical exam, automatically reporting the results to the administrator simultaneously. Accordingly, this research examines existing methods, identifies their limitations, and explores the significance of introducing a smart object-oriented program-based exam evaluator as a solution. This method minimizes all human errors and makes the system more efficient. An automated answer checker checks and marks are given as human-counterpart and generate a report with possible suggestions for improvement of the answer scripts and generate a classification report to predict the student’s final exam marks. This software application uses a Knowledge base, Abstract Syntax tree (AST), ANTLR, Image processing, and Machine Learning (ML) as key technologies. The proposed system gains a higher accuracy of 95% as performed by a separate human-counterpart. These results show a high level of accuracy and automate marking is the major emphasis to save human evaluation effort and maximize productivity.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 Science Zone : An Augmented Reality based Mobile Application for Science(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) De Silva, W.; Naranpanawa, P.; Hettihewa, U.; Liyanage, P.; Samarakoon, U.; Amarasena, N.In recent years, technology has rapidly developed, and it has provided many technological advancements for the field of education with an attempt to improve and overcome its limitations. Augmented Reality is among these latest technologies which support to improve learning environment around the world. It can bring education to a new level which can help students in many significant ways. In Sri Lanka, augmented reality is rarely been used for the purpose of educational enhancements. Therefore, it was decided to develop an augmented reality embedded mobile application for the G.C.E Ordinary Level Students in order to make it easy for them to learn Science with more enthusiasm and interest. This research has been used marker-based approach to transmit images or objects in the text book into the real-world scenes in order to create a more productive learning environment for the students. The first version of the application covers four main areas in the Science curriculum, such as; Preparation of Acids, Human Anatomy, Organization of Plants and Biosphere Cycles. Feedback for the application was taken from randomly selected ten science teachers and twenty grade eleven students and accordingly the application was further developed. Their feedback proves that the application would satisfy the common requirements of students, and it would be an immense support in scoring good results for science.Publication Embargo PatientCare: Patient Assistive Tool with Automatic Hand-written Prescription Reader(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Kulathunga, D.; Muthukumarana, C.; Pasan, U.; Hemachandra, C.; Tissera, M.; De Silva, H.Most people in the world prefer to be conscious of the medications prescribed by physicians. Especially, the importance of handwritten prescriptions is prodigious in Sri Lanka because they are widely used in the healthcare sector. However, due to the illegible handwriting and the medical abbreviations of the physicians, patients are unable to find the prescribed medication information. This research is an attempt to assist the patients in identifying the prescribed medicine information and minimizes misreading errors of medical prescriptions. When a patient uploads the image of a prescription, the system converts it into unstructured text data by using OCR and segmentation, then NER is used to categorize medical information from given text. According to the other research, some solutions exist in other domains for the above mechanisms. But they gave less accuracy when tried to apply for this research due to the domain specialty. Therefore, as a solution to overcome the above discrepancy this approach allows users to scan handwritten medical prescriptions and blood reports and obtain analyzed reports in medical history. Results have shown that this approach will give 64%-70% accuracy level in doctor's handwriting recognition and 95%- 98% accuracy in medical information categorization of the prescription format.Publication Embargo Smart Personal Intelligent Assistant for Candidates of IELTS Exams(2020 2nd International Conference on Advancements in Computing (ICAC), SLIIT, 2020-12-10) Senevirathne, S.S.; Fernando, G.U.D.; White, J.B.; Divyanjala, S.T.H.; Samaratunge Arachchillage, U.S.S.; Dias, D.P.N.P.Many IELTS candidates encounter problems at the examinations and majority of them are unable to achieve their goals even though they strive hard to accomplish their targets. Candidates strive to achieve higher band score in exams, but fail to achieve them due to the ignorance of prevailing weaknesses which have to be identified if they were to succeed. At present, IELTS seems to be the most demanding exam among applicants who are planning to embark their higher studies or migration purposes. Currently, there is no proper mechanism to assist candidates and generate an improvement plan by identifying the weaknesses of them. As a solution, Smart Personal Intelligent Assistant for Candidates Exams (SPIACIE) has been proposed to detect IELTS candidates’ weaknesses through an analysis of their answers. The SPIACIE assesses four components (Reading, Writing, Listening, and Speaking) in IELTS exams. This paper is specifically based on the Long Short-Term Memory (LSTM) network model used to analyze the score of grammar and cohesion. To analyze the similarity of the sentences, the cosine proximity technique is proposed to evaluate the paraphrasing of the graph explanations. The final outcome of this application is to generate an improvement plan, developed using Machine Learning (ML) algorithms. The proposed algorithms are; Gaussian naïve base for reading exam, support vector machines for listening exam, decision tree classifier for speaking exam, and k-neighbors classifier for writing exam. An improvement plan on the prediction model is provided to increase the band score of the IELTS exams, based on applicants’ weakness.
