International Research Symposium of UWU-2018

Permanent URI for this collection

https://erepo.lib.uwu.ac.lk/handle/123456789/1304

Browse

Now showing 1 - 20 of 22

An Accurate Multiple Sequence Alignment Algorithm for Biological Sequence Sets with High Length Variations
(Uva Wellassa University of Sri Lanka, 2018) Jayasingha, J.A.D.T.B.; Wannige, C.T.
Multiple sequence alignment (MSA) is used for many studies in modern biology. There are many algorithms available for the alignment of multiple sequences. Among them, progressive alignment algorithm is the most commonly used heuristic alignment strategy for MSA. It solves MSA with an economic complexity but does not provide accurate solutions, because there is a conflict between accuracy and complexity. The existing similarity score method in progressive alignment algorithm does not consider the lengths of the sequences in the considered sequence set. So, if the protein or DNA sequences are with high length variations, the initial alignment scores may not produce a correct measure of similarity between the sequences. This leads to less accurate initial alignment scores, and as a consequence, final multiple sequence alignment may produce inaccurate results. In this research, we present a modified progressive alignment algorithm especially for sequences with high length variations. We modify the latest version of ClustalW 2.1 by replacing the similarity distance measure in ClustalW algorithm with a novel distance measure. The new distance score method captures the distance between each sequence pairs in sequence set and the obtained distance measure is utilized to generate a better guide tree for progressive alignment. In order to determine the pairwise similarity distance measure, we used lengths of the shortest common super-sequence (SCS) and the Longest Common Sub-sequence (LCS). We assessed our algorithm with BALIBASE 3.0 protein benchmark and compared the obtained results to those obtained with ClustalW alignment algorithm using the Quality score (Q Score) and the Sum of Pairs Score (SPS). We obtained better Q scores and SP scores for the alignments from modified ClustalW algorithm over original ClustalW algorithm. Furthermore, the alignment speed of modified ClustalW algorithm is multiple times faster than the original ClustalW algorithm. Keywords
Analysing Critical Factors Associated with Perceived Risk for Major Types of Cancers in Sri Lanka — using Data Mining Techniques
(Uva Wellassa University of Sri Lanka, 2018) Parakrama, M.L.; Kodikara, K.A.K.D.; Senanayake, S.H.D.
Cancer is one of the world recognized critical cause of morbidity and mortality. Identification of demographic, health and life style factors are significant to detect and prevent from the cancers. Since the Sri Lankan public has lack of knowledge about influence of those factors on cancer risk, the research has been conducted using data mining techniques to develop significant patterns in order to identify the critical factors towards the most recorded cancer types in Sri Lanka (Breast, Oral and Oesophagus cancers). Hence, the ultimate objective of the research is to increase the awareness among Sri Lankan public which will be helpful to take preventive approaches. The research was based on the patients' data within a fiveyear period of time, which was collected from the sources of Apeksha hospital, Maharagama. By following the Knowledge Discovery in Databases process, data collection and preprocessing were completed manually. Data was analysed through the data mining tool called Weka. The research was used Random Forest classification technique, k-mean clustering technique, and Apriori association rule mining. From the research it has been identified that the gender, number of children, breast feeding, menopause, weight, age at menarche as the critical factors towards the breast cancers. The gender, marital status, weight, smoking, consumption of alcohol, betel chewing were identified as critical factors towards the oral cancers. The gender, consumption of alcohol, occupation, weight and age were identified as critical factors towards the Oesophagus cancers. The patterns were obtained from the above identified critical factors with their approximate values.
Artificial Conversational Agent Based Tour Guide System
(Uva Wellassa University of Sri Lanka, 2018) Bandara, S.B.G.J.S.; Jayasundera, J. M. K. H.; Udayanga, U. H. N.; Iloshini, P.A.A.; Pathirana, K.P.P. S.
The study designed to show that the tour guidance for tourists can be improved by incorporating all Artificial Conversational Agent (ACA). Although some of the aspects in Sri Lanka are well satisfied in the tourism industry, there is a lack of travel planning and collaborative interaction with the tourists. The implemented system is proposed to fulfil those solution gaps using a web portal with localized data and an ACA, who interacts with the user in natural language to provide information and trip planning functionalities. User expectations are identified by a statistical survey and the relevant dataset is based on localized destinations, intentions, locations, accommodations and route entities in Uva province of Sri Lanka for the Named Entity Recognition (NER) task. The ACA module is integrated with the web portal for users to interact using natural language. Dialogflow Natural Language Processing as a Service (NLPaaS) is used for Natural Language Processing (NLP) and Machine Learning (ML) tasks that are based on the custom entity model and an intents model, used for intent and entity extraction from user utterances. A web application implementing a webhook which is connected with dataset is used to query the data and an HTTP POST request is used to access the application endpoint and send relevant parameters. Optimized trip plan generation is implemented for certain user requirements and made it available to the user by sending an email. Training of ACA is done by Part Of Speech (POS) tagging for entities in potential user utterance corpus for each intent. For more user convenience the ACA is made available via social media channels and it is recommended to extend to the whole country. The effectiveness of the application is affirmed by the local and foreign tourists' feedback and MCQ answers, which are taken after using the application. It concludes that Sri Lankan tourism can be improved by efficient information delivering using emerging communication methods. Keywords
Artificial Intelligent Base Virtual Learning Assistant
(Uva Wellassa University of Sri Lanka, 2018) Vimalan, K.; Banusharma, U.; Jayasekera, N.E.C.
E - Learning is the most effective learning method that includes a variety of tools on learning styles and helps students become more versatile learners. Proposed system will encourage active learning of students in online learning and would be able to handle this problem efficiently. Further, whenever user searches for a problem, system will automatically search required information from knowledge base and provide a solution to overcome the problem when user needs, thus, it is a user-friendly system. Currently in Sri Lanka, there is no any specific virtual learning platform that connects all university students. Therefore, the major objective of proposed system was to develop a platform that will connect all university students together as an effective communication portal for academic matters. This System was equipped with a virtual learning assistant called Jarvis. Thus, Windows + IIS + ASP.NET + MSSQL combination was used for fast, reliable and efficient E-Learning web application and also Artificial Intelligence Markup language (AIML) was used for virtual assistant. Initially, a system was developed based on selected best technologies for efficient elearning application, then an Artificial Intelligence Markup language based virtual assistant was developed. Finally, all the components were bound to create a complete System. While the implementation of this system, the chatter bot answered the questions of students through speech. Consultants can teach their knowledge to Jarvis when necessary. The proposed system was successfully developed and also tested results complemented with the requirements.
A Conceptual Framework for Flood Early-Warning System for the Lower Flood Plains of Kalu Ganga Using Twitter Crowd-sourcing and Internet of Things
(Uva Wellassa University of Sri Lanka, 2018) Ranasingha, T.B.B.M.; Chathuranga, D.W.R.; Gunasekara, K.M.; Wimaladharma, S.T.C.I.
Flood is one of the common natural disasters in all over the world. Sri Lanka has two major monsoons: Southwest (May to September) and Northeast monsoon (December to February) causing for floods along the one third of low lands. The objective of this study is to develop a framework that is relatively credulous community based flood early-warning system for the populous areas near by the riverbanks of Kalu Ganga in Kalutara district in Sri Lanka. The study focuses on two major affected areas that are Palindanuwara and Agalawatta. There are six major tributaries joined to the river between Kalutara and Ratnapura making Ratnapura as a considerable catchment area for the river. Therefore, the system collects real-time bulletins, associated with predefined keywords and posted by the Twitter crowdsourcing living in Ratnapura and surroundings, using Twitter stream API. It uses hashtags to filter locations and performs the text analysis. While the percentage of likelihood of flooding is estimated based on the number of positive twitters, the possibility of a flood is verified using the incline or decline trend of the water levels collected from Ground Control Units located in flood risk areas. If the Ground Control Unit confirms that there is a possibility of a flood, the system generates a flood-positive alert that can be used to warn people living in those areas. The proof of the concept was successfully tested by simulating the flood situation using the Ground Control Units. Thus, it can be concluded that the Twitter crowdsourcing can be effectively used to warn the community about upcoming flooding situations beforehand.
Deeper Emotional Capture and Extraction of Discussion Pathways in Twitter Data
(Uva Wellassa University of Sri Lanka, 2018) Epaliyana, K.V.; Jayatissa, K.N.A.; Liyanage, A.J.H.; Sandeepa, K.G.L.
The rise of micro-blogging has resulted in people expressing their daily thoughts, often resulting in far more emotion-laden than might normally occur. Recently, it has become a trend to post about the places one has visited and experiences the person had. Finding the emotions expressed through these texts can be used in understanding community thinking and decision making, but it is quite complex since it requires thorough knowledge in psychology and linguistic. Furthermore, processing the microblog text is challenging, because they are informal and less consistent in terms of language. This paper presents a lexicon and rule based approach, which breaks the emotions behind a tweet into the eight basic categories a person is capable of expressing, as defined by Robert Plutchik. Hence, we broke the emotional tone behind a tweet into 8 basic emotions namely anger, anticipation, disgust, fear, joy, sadness, surprise, and trust. In this, we selected a tweet corpus related to Sri Lankan tourism by querying a Twitter tool with suitable keywords often used by tweeters. We have augmented the accuracies of emotions capturing through a series of extensive text preprocessing steps fitting to twitter texts. The experimental results have shown that processing the informal tokens, hashtags and repeating character sequences can make a significant improvement in the emotion capturing accuracies. This research includes separating the tweets into cohesive discussion pathways using an existing incremental unsupervised machine learning algorithm, in order to come up with a rich decision making tool for the selected domain. No previous research has broken down emotions into these eight basic emotions and integrated the separation of discussion pathways into emotional capturing. We have contributed to the research world by succeeding in both these untapped research areas and developing a user friendly tool to use in constructive decision making.
DIGITAL FORENSIC STEGANALYSIS OF ENCRYPTED INFORMATION WITH SPECIAL REFERENCE TO THE MP4 VIDEOS
(Uva Wellassa University of Sri Lanka, 2018) Palliyage, S.L.; Kankanamge, N.D.; Ariyadasa, H.M.S.N.
Computerized information correspondence has turned into an integral piece of foundation these days and the security and the privacy of them assumes an essential part. Cryptography and Steganography are being two major methods having a wide usage, which secure information. Steganography conceal secret information and does not leave evident proof of information modification. Out of numerous techniques of application of Steganography, current study focused on End of File Injection technique. The study investigated the Steganalysis of Mp4 type videos, which were encrypted utilizing the above technique. A pool of Mp4 videos consisted of several qualities and capacities were tested and evaluated. A detector system was developed which was capable of identifying the presence of Steganography within the Mp4 videos, thus the system can be further used to scan the suspected Mp4 videos and give the results whether it has embedded data or not. The system reported a higher accuracy level of detecting Mp4 Stego-videos, which used the said techniques for data embedding. Further studies are needed to cover other video formats and other techniques of Steganography. The development of the field would reveal new paths in digital forensic investigations.
Dynamic Offloading Algorithm for Cyber Foraging
(Uva Wellassa University of Sri Lanka, 2018) Wijeratna, J.A.S.; Karunarathne, K.M.C.M.; Priyadarshani, G.A.N.; Nilwakka, H.I.
Advances in semiconductors and computing technologies have enabled small consumer devices such as smart mobile devices to perform heavy computations. Also, the mobile applications we use today are highly resource demanding. Though the mobile devices are capable of running heavy tasks their resources are limited compared to desktop computers or cloud computers. So, we can use the technology called Cyber Foraging which enables mobile devices to use cloud resources to perform resource demanding tasks. In this research paper, we are presenting an efficient algorithm to take decisions dynamically about offloading heavy tasks from mobile device to a resource rich cloud platform. We have identified the project named Rapid, which is the best opensource framework for cyber foraging so far and we implemented an algorithm on top of that project. In the Rapid framework offloading process is done in a static manner by annotating identified resource intensive methods at the development phases. Our algorithm considers the situational factors such as network bandwidth and battery level of the mobile device to take the decision. Also, it considers the previous execution details such as average power consumption and average execution time for both remote and local executions. If the battery level is high and power consumption and execution time for certain method is lower, then that method will not be offloaded to the cloud. If the battery level is lower than a threshold value and execution time is higher for certain method, then that method will be offloaded. We have tested our dynamic approach using n-queen algorithm. According to the results the decision-making algorithm helps resource poor mobile devices to execute resource intensive tasks efficiently and thus provides better and smooth user experience.
An E-Management Solution for the Sri Lankan Police Department to Improve the Efficiency of Service
(Uva Wellassa University of Sri Lanka, 2018) Alahakoon, A. D.C.; Abeywardhana, K.G.H.; Karunarathna, S.S.W.; Senanayake, S.H.D.
Lack of information sharing mechanism, duplication of data and delay of services were identified as the main problems in the Police Department of Sri Lanka. Therefore, an e-management solution was introduced to fulfill the above requirement. The efficiency of services in the Police Department will be improved by this e-management solution. Better and efficient services will be provided to the victims as well as police by automating the current manual process. This solution has two parts named web system and the mobile application. The computerized web system will facilitate complain handling and provide reports by printed documents and keeping criminal records on them. The records can be stored and retrieved more securely through this web system. The mobile application was introduced to communicate between police officers in each division in the Police Department very easily and fast. This application mainly consists of parts such as check people identification, check vehicle, add accident, add fines and message. Details of people and vehicle can be verified by police officers, who are out of the police station and messages can be exchanged more securely, fast and clearly by police officers through this application. The web system and the mobile application were connected by using web server through Json API and MySQL server was used to create the database of the research. Efficiency of police performance and procedures can be improved by this e-management solution through integration of previously separates information systems.
An Intelligent Postal Mail Sorter: Sinhala Hand Written Address Recognition Method Using Geometric Feature Extraction Technique and Artificial Neural Network
(Uva Wellassa University of Sri Lanka, 2018) Sri Darshana, B.P.S.R.; Attanayake, A.M.U.L.; Perera, A.A.L.A.C.; Wimaladharma, S.T.C.I.
The main objective of this study is to develop a methodology to recognize Sinhala handwritten characters that can be used in postal mail sorting. The Department of Posts, Sri Lanka uses the manual sorting mechanism, while most of the developed countries are using automated sorting machines. The main reason for not having such types of machinery in local postal collecting and distribution centers is the initial cost of implementation. The machines have to be tailor-made due to the localized language. The proposed methodology is based on the geometric feature, such as Corner detection, Curve fitting and Edge detection, extraction technique and Artificial Neural Network backpropagation technique. The benchmarking of the classification system is carried out using 34 Sinhala characters that are mostly related to the district names. The neural network consists of three layers, where the input layer with 108 input nodes, the output layer with 34 nodes and a hidden layer of 78 nodes. The training and testing are performed by 850 characters and 510 characters, respectively. The accuracy of the system is around 78% of giving a correct answer. The resultant set of characters then be extracted and used to control the sorting machine. In order to prove the concept, an embedded system is developed using Arduino microprocessor. The sorting mechanism is simulated by using a servomotor that indicates the relevant mail bucket using a rotating arm.
Location Based Exploratory Decision Support Approach for Midwifery and Grama Niladhari Divisions in Hunuwala-North: Ratnapura District, Sri Lanka
(Uva Wellassa University of Sri Lanka, 2018) Wickramasinghe, W.B.P.L.; Ariyarathna, K.L.G.A.M.; Wimaladharma, S.T.C.I.
Spatial information is being used as a supportive component for the process of decision making in various disciplines and applications. Generally, the governing activities, facts of citizens and properties, and natural and man-made phenomena are associated with locations. In Sri Lanka, the smallest administration unit is the Grama Niladhari division, whereas the midwife is the closest health care person that the community encounters. Most of the time two divisions are overlapped with each other resulting that they collect common facts about citizens. All the data about villages or citizens collected by Grama Niladharies and midwives become the data sources to make decisions by the top-level officers. The main objective of this study is to develop a location based (spatial) decision support approach for multiple criteria decision model with geo-visualization for decision making officers in various government sectors such as divisional and district secretaries, top-level officers of healthcare sector and their upper administrative levels: Its architecture consists of three major components namely spatial layer, attribute layer and the criteria layer. Each attribute record is associated with at least one spatial record resulting to a geospatial database, which has citizen level data, with predefined rules and criteria compiled according to the administrative policies and healthcare rules and regulations of the government. A proof of concept is developed and tested with the actual data. Therefore, it is proven that the introduced approach has a significant effect for the decision makers to make cognitive decisions rather than emotional decisions.
A Novel Method to Measure The Water Content of The Leaves Using Digital Image Processing
(Uva Wellassa University of Sri Lanka, 2018) Wagachchi, I.A.; Rassagala, R.D.K.; Abeykoon, B.B.D.S.; Kartheeswaran, T.; Jayathunga, D.P.
Leafy product industries like Tea, Tobacco, Palmyra, Leafy vegetables, and Ayurveda productions play a significant role to uplift the Sri Lankan economy. The water content in the leaves is an essential factor for leafy productions to maintain their quality. Naked eye observation of an expert is the general method to identify the water content. The objective of this study is to introduce a novel and easy method to measure the water content of the detached plant leaves using digital image processing. As a result, a simple computational water content prediction method has been built using image processing techniques to obtain a quality output at the end of production processes. The findings of this study help to identify the water content without an expert in an efficient manner. First, the colour images were captured in a control environment, while leaves were drying and simultaneously the weight was measured traditionally to find the water loss. Features were analysed from images to find the best features, which show a better correlation with the changes of the water content in the leaves. The basic features such as homogeneity, energy, contrast, variance, mean, median, min, max, range, kurtosis, skewness, standard deviation, entropy, correlation and IQR were extracted. The best features among the selected features have been chosen through correlation test. The classification was done with the K-Nearest neighbour algorithm by training with the selected best features of the training set of images. The green matrix of the RGB image is taken for the feature extraction to get better results. Finally, a simple model was built using the significant features which have a relationship with the water content measurement. 65.3% accuracy has been achieved, and this model can be used to predict the water content of a particular green leaf through images. This model will be a turning point for measuring the water content of the leaves in the industries in an automated manner.
Online Event Organizing System for Cultural Programme
(Uva Wellassa University of Sri Lanka, 2018) Shobika, G.; Yogaseelan, Y.; Ramanan, M.
This paper proposes to develop an Online Event Organizing System for cultural programme which can provide various event services and give efficient coordination between customers and service providers. People are tired of searching and travelling to find quality and economical event services as they want their traditional larger events, attractive and interesting. Using the existing manual system, an event planning is stressful, time consuming, service providers reputation is unconvincing and customers do not know what to look for in an event service providers. The proposed method consists of establishing requirements, analysis, database design and interface design. The proposed work is a benefit for the users and as well as for service providers. It provides opportunities to service providers advertise and develop their business and show their interesting quality ideas. An event organizing system makes the people easier to plan an event via online. The proposed work is developed an online event organizing system with a mobile front end to manage various events for a cultural programme. The system is developed using HTML, JavaScript and PHP.
Optimizing Boyer Moore Algorithm using Parallel Processing Techniques
(Uva Wellassa University of Sri Lanka, 2018) Pathiranage, S.Y.D.; Ediriweera, T.S.; Thambawita, D.R.V. L.B.
Although computer processing power, memory and parallel processing capabilities have been significantly improved, most of the algorithms are still not improved based on novel technologies. Boyer Moore string searching algorithm is one of a mechanism that uses frequently. Thus, we optimized Boyer Moore String searching mechanism using parallel computing mechanism. Existing implementations were analyzed, compared the performance using CPU time with different pattern sizes in same condition. Python parallel version was implemented using three threads. Serial Python code uses main thread for two shifting rules calculation and searching mechanism. Thus, we used two threads for shifting rules calculation and one thread for searching mechanism. Existing implementations used one way searching mechanism to search the pattern in the text field. Then searching mechanism was changed into bidirectional searching process using a parallel Python implementation of the Boyer Moore Algorithm with six threads. To optimize the algorithm than bidirectional searching parallel python version, processing logic of the Algorithm was changed and used four threads. Two threads for the sifting rule calculation and after those two threads finished, string text is divided in to two parts and search begins with two threads for each part. Existing serial python version took 165.44 seconds for found correct pattern. However, our first implementation took 163.33 seconds for found correct pattern. Because of bidirectional searching way, the second implementation took 320.94 seconds,. Third implementation took 100.98 seconds for found correct pattern. Measure the time in the same conditions with the same text (T) and same pattern (P). Time is depending on text (T) size, pattern (P) size and machine type that we use to run the codes. Final implementation is bidirectional searching parallel python version with four threads. It is 38.96% effective than existing serial python version.
A Personalized Content-Based Recommendation on Knowledge Level of Learner
(Uva Wellassa University of Sri Lanka, 2018) Herath, D.; Jayaratne, L.
e-Learning is considered as one of the most popular research areas in distance and web based education. Nowadays, most of the education institutions such as universities, colleges and vocational training centers are adapted e-Learning environment to give quality and efficient service to learners. This paper presents a novel approach, a framework for building a personalized content-based recommendation system by considering learner's knowledge has been developed. The knowledge, skill, preferences and learning style of each learner is different. Therefore, we should understand different needs of learners and provide a better recommendation to motivate them. The proposed recommendation system consists of four components, Learning model, Domain model, E-Assessment model and Recommendation model. In e-assessment model, the learner takes diagnostic, formative assessments in different levels such as initial, final, practice and assignment. While learner attempts, the system generates recommendations based on level of knowledge. Through the system, each learner's progress can be identified and compared with other e-learners' results. Then, the learner motivates to learn more learning objects and stick to self-learning style. During this research, Content-based filtering is used as filtering approach for making recommendations. Apart from that, web mining techniques has been used to implement result predictions, clustering e-Learners and classifying learning objects. However, cold start problem has been minimized by using an initial test at the start point of each learning module. Furthermore, the impact of introducing the e-assessment model to support and improve the learning process is evaluated following the action research methodology and comprised two main activities: system testing and validation, in a real scenario, a programming course in Bachelor of Information Technology at University of Colombo. In this research, decision support accuracy metrics was used for evaluating recommendation algorithm. A set of students who is following above course was selected to determine knowledge levels and performances. Based on data analysis, it was observed that the use of the system and the models had a positive impact on student learning and performance. Students learned through more interaction with the system and, as a result, their performance in the assignments and final examination had improved. The system also provided added benefits to facilitators for tracking of students' progress throughout the whole course.
Preliminary Study on ICT Applications in Agriculture to Enhance Information System using Mobile Crowdsensing
(Uva Wellassa University of Sri Lanka, 2018) Wilson, R.S.I.; Goonetillake, M.D.J.S.; Walisadeera, A.I.
This study analyzed the existing Information and Communication Technology (ICT) applications in the agriculture information system and proposed a method to enhance a mobile-based information system using mobile crowdsensing. The proposed system enables farmers in Sri Lanka to report events such as diseases, about their crops and get advices for making farming decisions. The lack of realtime information environment makes issues in the farming life cycle and that will affect the national economy and employment. Existing studies were analyzed for identifying the capability to establish a mobile-based information environment in the farmer-community in Sri Lanka. The applications introduced in the studies are covering several aspects such as market price, vendors, crop details, pest and disease information, etc. In order to create a real-time information environment, the real-time data related to farms need to be collected. Mobile sensing is a technique to obtain real-time data from a large group of individuals. Thus, the mobile sensing technique introduces for the farmer-community that allows farmers to participate the system by sharing mobile sensing data like images, text, voice, location, date and time. The knowledge-base of the system contains knowledge about crops, diseases, pests, fertilizer requirements and control methods for growing problems. Thus, the agriculture information in the knowledge-base can be accessed by the farmers according to the shared sensory data. Over the time period, agriculture information can be updated as the technology advancement, seasonal changes or unexpected weather changes. Thus, new knowledge needs to be stored in the agriculture knowledge-base with the help of domain experts. Moreover, by reasoning the collective data shared by farmers and experts, interesting aspects like the suitable crops to grow in a particular area, the crops badly affected by a specified disease in a particular area or treats rapidly spread in a particular period, etc. can be obtained for decision making. Further analysis can be done for future predictions from the large collection of data.
Process re-engineering of Sri Lanka's Downstream petroleum industry stock planning operations.
(Uva Wellassa University of Sri Lanka, 2018) Dharmaprema, J.G.J.W.; Gunasekara, W.V.
The vision of the Ceylon Petroleum Storage Terminals Limited (CPSTL) is to be the most efficient petroleum terminal operator in South Asia. CPSTL is responsible for 90% of storage and distribution of the Sri Lankan downstream petroleum industry, which constitute of tanker/tank farm operations, island-wide delivery of petroleum products by rail, road, and pipeline. CPSTL provides ERP and laboratory services for the entire industry. The logistical operation is accomplished via 2 main installations and 12 island-wide bulk depots. It involves fuel supply to dealers and consumers, whilst replenishing the island-wide bulk depots. The installations are replenished by tankers from Colombo harbour, Muthurajawela offshore SPBM and the Ceylon petroleum Corporation's (CPC) Sapugaskanda refinery. Bulk depots are replenished by the 2 installations via tank lorries and railway wagons. Although stock planning procedures did exist, various limitations and drawbacks have resulted in underutilization of resources, massive losses in distribution operation, increased risk of storing highly inflammable products and window for various manipulations. The project was implemented with the objective of streamlining the stock planning operation with the utilization of the SAP ERP system. It provides a daily stock quota system for each bulk depot, which optimizes the tank lorry, wagon utilization as per the sale, stock availability, tank capacity, goods receipt status, and vehicle availability. Two important phases of stock planning operations were identified during business process analysis. They were re-engineered and integrated to SAP ERP system. The system ensured island-wide product availability eliminating over/ under scheduling. The system included features for tracking of delayed tank lorries and wagons. Various key performance indicators related to the stock planning operation were introduced to facilitate operational monitoring and strategically important decisions.
Recognition of Vowel Distribution for Sri Lankan Traditional Pirith Chants Using Formant Variation
(Uva Wellassa University of Sri Lanka, 2018) Gunawardana, M.A.C.P.; Gamage, S.S.N.
Pirith is believed to be a doctrine preached by Lord Buddha and regarded to obtain protection from evil, promote health and well-being. Voice source produces a harmonic series, consisting of the fundamental frequency, F0 and harmonic frequencies called as formants, Fn. Vowels can be mapped using the relationship between lip opening width to the first formant frequency, F1 and tongue constriction width to second formant frequency, F2. This work is dedicated to developing phonetic picture on Pirith chants and analyze acoustic properties using computeraided tools. As reported by several other studies, characteristic vowels and high frequency formants are identified in chanting in contrast to normal speaking. The motive of this study is to investigate vowel distribution of Pirith chants with the aim of special pattern recognition. Samples of Ratana, Karaniya Metta and Angulimala Suttas recited by male monk chanters were recorded using high precision microphone array and 15 samples of each Sutta were analyzed. Recorded samples were then subjected to splitting of smaller voiced segments of frame length 10 ins using sampling rate of 44.1 kHz. In the computational speech model, a preemphasis filter is applied to the sampled time series of voiced segment to cancel out the effect of glottis. Then frame-by-frame analysis was used with hamming windows and liner predictive coding (LPC) and auto correlation to extract the formant values. Finally, PDFs of each Sutta is generated and compared for first five formants. Angulimala Sutta and Ratana Sutta show similar patterns in terms of PDFs but Karaniya Metta Sutta indicates a clear discrepancy demonstrating a unique set of characteristics. Furthermore, the vowel distribution reveals that Angulimala Sutta and Karaniya Metta Sutta contain high number of compact vowels comparing that of Rathana Sutta.
Resource Creation for English-Maithili Machine Translation (EMMT) A Divergence Perspective
(Uva Wellassa University of Sri Lanka, 2018) Nidhi, R.; Singh, T.
Maithili is one of the 22 scheduled Indian languages with almost no language technology resource. Absence of basic tools in this language has affected resource creation. Since English is the dominant language, translation from it can help creating the required corpora for tools development in Maithili. The present work discusses efforts for Language Technology Resource (LTR) creation and divergence study for an EMMT system, which is a Statistical Machine Translation (SMT) system. Creating any SMT system requires sizeable parallel, aligned corpora for training and testing. Creating general-purpose source corpora for English language and creating translation equivalents with possible alignments requires time and effort. The paper focuses on the data collection methods, cleaning, the size and structure of the text corpora, alignment and parallelization strategies, training, testing and a study of divergence between the language pair.
Sixty Seconds Robot - An Effective Development Approach for Robot Programming
(Uva Wellassa University of Sri Lanka, 2018) Priyanga, E.A.I.; Jayarathna, H.M.P.P.; De Silva, B. P.; Senanayake, S.
Nowadays, a lot of people are interested in robotics technology. However, most of them do not have a required knowledge in robotics and pogramming. Therefore, it has become a major barrier for innovations in robotics. However, if the basic process is automated, the entire problem can be solved without wasting programmers' effort and time. There are some visual programming software existed for programming in robotics such as Visuino, GLTINO, XOD.IO, etc. The main problem of those types of advanced software is that the learning difficulties at the beginner level. Some of those tools are not much user-friendly for novice learners. Some of them do not facilitate to identify hardware related errors in the robot. To overcome these barriers, an efficient software tool has been introduced in this research to design the basic structure of a robot. The automated IDE can identify hardware related errors. There are many types of robots, such as line follow, wall detect, arm based, etc. The user can select a template from the IDE for one of the types and customize it. Also, it can be used for tuning sensors, Proportional Integral Derivative (PID) generators, etc. In this solution, the user has only to customize the template, and then the code for Arduino circuit that is generated automatically. Since, this is a full GUI application to do Arduino robotics, this provides a platform for the beginners to learn robotics and make their own projects with less errors. Our future development is mainly focused on developing solid work based on a universal template for more Artificial Intelligent (AI) based robots, radio signal based robots, etc. Ultimately, those extensions can be used in simple applications to more advanced applications such as robot based toys, military tools, drone technology, etc. In addition to that, the researchers have planned to develop a community for that where anyone can create libraries for implemented robotics platforms.

Browse

Browsing International Research Symposium of UWU-2018 by Subject "Computing and Information Science"

Results Per Page

Sort Options