ICITR - 2018

Permanent URI for this collectionhttp://192.248.9.226/handle/123/14730

Browse

Now showing 1 - 20 of 34

item: Conference-Full-text
3rd International Conference on Information Technology Research 2018 ( Pre Text)
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Wijesiriwardana, CP
item: Conference-Full-text
Predictive maintenance and performance optimisation in aircrafts using data analytics
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Weerasinghe, S; Ahangama, S; Wijesiriwardana, CP
Airline industry has provided a significantly conventional, faster and reliable mode of transportation for passengers and freight over the decades in which the industry has been in service despite the pressure being applied especially in maintaining operational affordability. The study critically reviews the techniques and tools, infrastructure and general application architecture for discussing the applicability of data analytics based on both batch processing and real time stream data in general aviation for health monitoring and predictive analysis in order to predict maintenance and optimize the performance of aircrafts. In this respect, the study further evaluates the significant capability in addressing contemporary problems which are uniquely addressed by data analytics system.
item: Conference-Full-text
A review of big data analytics for customer relationship management
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Perera, WKR; Dilini, KA; Kulawansa, T; Wijesiriwardana, CP
Big Data Analytics is a major research topic in the business world and maintaining a good Customer Relationship Management is a major requirement. Handling Big Data according to Velocity, Volume and Variety is a major issue. There are many challenges in using Big Data in Customer Relationship Management. To include Big Data in CRM many techniques such as Data Mining, Frameworks and many procedures are used. At present there are many applications in Big Data in customer Relationship Management and there are many limitations in those systems. This research paper attempts to analyze researches of Big Data Analytics, Data Mining techniques, Big Data Analytical Frameworks that can be used in Customer Relationship Management. By analyzing these researches this paper describes current applications of Big Data in Customer Relationship Management, their issues, limitations and future directions of this field.
item: Conference-Full-text
Forecasting stock price of a company considering macroeconomic effect from news events
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Waduge, N; Ganegoda, U; Wijesiriwardana, CP
Investment perspective, Stock Market is the most popular potential investment market around the globe currently, because of this reason the need of an effective stock prediction approach was a target of many researchers. Most of the previous approaches have adopted Artificial Neural Networks and Support Vector Machine, while some have got insights from other models as well. Above traditional approaches was lacking in precise predictions of stock price fluctuations. This paper reviews the previous approaches with different Machine Learning methods and suggests a predicting method using modified Artificial Neural Networks with consideration of macro-economic effects to promise better results in stock prediction.
item: Conference-Full-text
-Blind draw-a software solution for image identification and artistic skills for visually impaired people using braille
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Fonseka, ODS; Wedasinghe, N; Wijesiriwardana, CP
At present, visually impaired individuals do not perform or engage much in art. They do not draw paintings because they do not have the ability to draw a world unknown and unseen before. New emerging softwares are developed within the world and the success of it is enjoyed only by the visual impaired who can afford it. The conversion of image to braille and getting a braille printout of the image is the main aim of the new software application. By studying previously developed systems, new features were identified. Reducing the complexity of the software solution and providing the main output of a well converted image from the basic shape to complex image conversions will be made possible. With the results of the survey conducted for the research it emphasized the necessity of inventing a software solution to give the opportunity to the visual impaired to engage in art using braille.
item: Conference-Full-text
Defining of normalized load profile curves for domestic customer groups to estimate feeder power loss
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Jayawardhana, HACH; Hemapala, KTMU; Bandara, AWAL; De Silva, PSN; Wijesiriwardana, CP
In a country, domestic electricity customer percentage is higher in number wise, but energy usage of one customer is lower compared to other categories. Therefore, installing a load profile recording meter for each domestic customer is not worthwhile and impractical. In this research, a methodology is proposed to estimate domestic customer load profile by using customer information to avoid the use of advanced costly energy meter. In methodology, the domestic customers were divided into several groups by clustering their daily load profiles according to differences of patterns. Representative normalized load profile is defined for each group. Same customers were interviewed for collecting family member information and electric equipment usage information. Relationships between load profile and customer information were investigated. Then a methodology was developed to estimate load profile of another new customer. These load profiles were used for calculation of low voltage feeder power loss estimation. As outcome of this research, MATLAB GUI software interface was developed to input customer information and selection of relevant load profile of a new customer depending on customer information. An algorithm is proposed to estimate hourly LV feeder power loss variation by using preestimated customer load profiles.
item: Conference-Full-text
Survey on wireless sensor networks (wsns) implemented for environmental sensing
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Galappaththi, HR; Weerasuriya, GT; Wijesiriwardana, CP
In Wireless Sensor Networks (WSNs) a large number of wirelessly-connected sensor motes work in collaboration to achieve a common goal such as detection of earthquakes, tracking animal movement, monitoring infrastructure or a volcano, and process control. Among the deployed WSNs, environmental sensing is important as a considerable number of the WSN applications have been developed on environmental sensing. The collected environment sensor data such as humidity, temperature, and air quality and soil moisture can be analyzed for better decision making. Moreover, actuator systems can be operated based on the sensor data. This paper presents a survey of already deployed WSNs for environmental sensing, comparing their scale of deployment, topologies and architectures, and routing protocols.
item: Conference-Full-text
Google map and camera based fuzzified adaptive networked traffic light handling model
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Nirmani, A; Thilakarathne, L; Wickramasinghe, A; Senanayake, S; Haddela, PS; Wijesiriwardana, CP
Rising traffic congestion has turned into a certain issue as the number of vehicles on roads are increasing. This research study was conducted to develop ‘Google Map and Camera Based Fuzzified Adaptive Networked Traffic Light Handling Model’. The main road with six major junctions was selected as the target route for the project. During this study, we were able to plan a limit and control traffic congestion utilizing two neural networks which process together to provide an efficient, productive and optimized solution based on real-time situations. Real-time video streams and Google Map traffic layer were used as primary input sources to the system. The Main algorithm was used to reduce traffic at a specific point whereas secondary algorithm was used to produce optimum decisions for the overall network. As a further advancement, REST endpoint was implemented to get the best route considering all the accessible data. With the aid of the previously mentioned techniques, an optimal traffic management model was developed.
item: Conference-Full-text
Alexza: a mobile application for dyslexics utilizing artificial intelligence and machine learning concepts
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Rajapakse, S; Polwattage, D; Guruge, U; Jayathilaka, I; Edirisinghe, T; Thelijjagoda, S; Wijesiriwardana, CP
Dyslexia can be explained as a neurological learning disability which causes difficulties in reading, word decoding, comprehension, short-term memory, writing, spelling, and speaking. People who are diagnosed with dyslexia tend to show signs of low self-esteem and anxiety since they can't interact with the society in a way that their peers do. Many applications available in this domain help them by correcting their issues by playing games and reading some hard-coded texts or pdf books. This correcting process takes time and dyslexics become helpless when coping with their day-to-day activities. This paper describes results of an evaluation of a prototype mobile application which helps the dyslexic users to deal with their reading difficulties in real life successfully, while they are receiving proper treatments. This prototype can identify the texts around them and read it loudly so that user can understand and will be allowed to customize the chunking, scrolling and highlighting of words according to their disability levels. By integrating dictionary support with the phonic and morphological structure of the word, the user will be able to comprehend difficult and complex words easily. In addition, the study also explores the use of a machine learning approach to improve the effectiveness of the learning dyslexic complex words.
item: Conference-Full-text
N-dimentional data visualization for industrial power consumption
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Samaranayaka, JRACP; Wimalaratne, P; Wijesiriwardana, CP
In the modern world, organizations will generate various types of data as a result of their day today activities. Collecting the data itself will not enough to take strategic advantage over them. Improved analytical techniques or applications will lead to take the maximum out of collected data over the competitor in the field. In our novel concept, knowledge of one variable will not simply depend on one or two independent variables; we have selected the dimensionality of the data as the most interesting parameter. Prototype of our novel concept is a web based approach which facilitates remotely conducting data visualization through a web browser without requiring any additional installations or plugins in the client side. One of the main features of our concept is less navigations/interactions which lead to minimize memorizing to mine interested patterns over data. Prototype of our concept shows increased usability for commonly all users over 2D visualizing techniques.
item: Conference-Full-text
Vision based office assistant robot system for indoor office environment
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Diddeniya, SIAP; Adikari, AMSB; Gunasinghe, HN; De Silva, PRS; Ganegoda, NC; Wanniarachchi, WKIL; Wijesiriwardana, CP
This paper presents an office assistant robot that can be used in an unstructured indoor office environment. Among many technologies available, we used free and open source software and inexpensive sensors and materials to build the low cost but accurate robot. Robotic Operating System (ROS) indigo was used as the ground operating system on Ubuntu 14.04. The mobile robot, iRobot Create 2 was used as the basic robot and a structure was built to carry a mini-laptop and PrimeSense 3D vision sensor. A workstation computer was used as the central location PC which was kept still and map of office environment is built on it. User is able to give commands to the system via voice and virtual keys by the developed Android application (App). These three units, mobile robot, central workstation and Android devices, communicates through a Wi-Fi connection. The proposed robot could deliver documents or parcels in between office members according to the user commands. We allowed the robot to navigate autonomously and randomly between users and monitored its accuracy by looking at the completion of the route to a target user. Results show that the office assistant delivery robot has above 92% of accuracy in delivery process for a valid user input.
item: Conference-Full-text
Augmented reality based breadboard circuit building guide application
(2018) Thiwanka, N; Chamodika, U; Priyankara, L; Sumathipala, S; Weerasuriya, GT; Wijesiriwardana, CP
Building circuits on breadboards is an activity which requires a lot of attention and thinking. If there is a way to guide this process by using modern technologies, the learning process can be made more effective and interactive. This study proposes a solution that provides students with an augmented reality visualization of the expected circuit on a breadboard before they actually make the circuit. The proposed system can be divided into four main modules based on their functionality (a) extracting possible information from the electronic components, (b) scanning circuit diagrams for identifying circuit symbols and their connectivity, (c) finding the appropriate arrangement of the electronic components on the breadboard and (d) using augmented reality to visualize the circuit on a breadboard. This solution provides an innovative approach to facilitate the learning process of students by making electronic circuit building interesting and interactive.
item: Conference-Full-text
Word level language identification of code mixing text in social media using nlp
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Shanmugalingam, K; Sumathipala, S; Premachandra, C; Wijesiriwardana, CP
Understanding social media contents has been a primary research topic since the dawn of social networking. Especially, contextual understanding of the noisy text, which is characterized by a high percentage of spelling mistakes with creative spelling, phonetic typing, wordplay, abbreviations, and Meta tags. Thus, the data processing demands a more complex system than traditional natural language processors. Also people easily mixing two or more languages together to express their thoughts in social media context. So automatic language identification at word level become as necessary part for analyzing the noisy content in social media. It would help with the automated analysis of content generated on social media. This study uses Tamil-English code-mixed data from popular social media posts and comments and provided word level language tags using Natural Language Processing (NLP) and modern Machine Learning (ML) technologies. The methodology used for this system is a novel approach implemented as machine learning classifier based on features such as Tamil Unicode characters in Roman scripts, dictionaries, double consonant, and term frequency. Different machine learning classifiers such as Naive Bayes, Logistic Regression, Support Vector Machines (SVM), Decision Trees and Random Forest used in training and testing. Among that the highest accuracy of 89.46% was obtained in SVM classifier.
item: Conference-Full-text
Thamizhifst: a morphological analyser and generator for tamil verbs
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Sarveswaran, K; Dias, G; Butt, M; Wijesiriwardana, CP
ThamizhiFST is a Morphological Analyser and Generator (MAG) for Tamil. It was developed to extend the coverage of the computational Tamil grammar being developed using Lexical Functional Grammar (LFG). ThamizhiFST covers the simple verbs in Tamil as an initial step. A Finite State Transducer (FST) approach was used to develop the MAG and it was implemented using the FOMA Open Source Software. Since morphological rules are of a finite nature and represent a known quantity, a rule-based approach like FST is more appropriate than possible machine learning alternatives, especially with respect to achieving reliably good accuracy that is required for computational grammar development. A set of 3250 Tamil verb lemmas from 13 paradigms together with their 260 conjugation forms were used in the construction of ThamizhiFST. Further, a set of 27 labels were used to mark the morphosyntactic information of the verbs. The whole system was developed as a three-layer web-based system to tackle the issues arising when processing an agglutinative language like Tamil and to ensure its extendability. Unlike other existing MAGs, ThamizhiFST also provides the morpheme corresponding to each morphosyntactic label and marks morpheme boundaries. An evaluation shows that ThamizhiFST has an f-measure of 0.97 for simple verbs. Future and current work include work on extending the system to cover more verbs and nouns and make it generally available.
item: Conference-Full-text
Building an open-source environmental monitoring system - a review of state-of-the-art and directions for future research
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Sudantha, BH; Warusavitharana, EJ; Ratnayake, GR; Mahanama, PKS; Cannata, M; Strigaro, D; Wijesiriwardana, CP
The massive development take place in the field of IoT (Internet of Things) has enabled generation of large set of real time environmental data from low cost sensor networks. This has enabled developing nations to monitor their environments with more economical options. The environmental monitoring network of Sri Lanka is mainly handled by the government bodies such as Meteorological Department, Irrigation Department and Central Environmental Authority. Most of the systems are manually operated and sparsely distributed. The automated systems are restricted to certain main cities of the country. Thus, the data provided by the existing network not capable enough to generate reliable climate or disaster specific forecasts. The available data is mostly retained as hard copies and they are not freely and publicly available. Hence, every time, the researchers and decision makers need to purchase data and required to undergo redundant data collection procedures. In this context, the 4ONSE research project on developing an experimental, open-source, low cost and non-conventional Environmental Monitoring System (EMS) can be considered as the 1st initiative in Sri Lanka which creates a new revolution in the country for environmental data viewing and sharing without any charge. 4ONSE is the acronym for 4 times Open and Non-Conventional technology for Sensing the Environment, which is a joint research between University of Moratuwa and University of Applied Sciences and Arts of Southern Switzerland (SUPSI). Deduru Oya river basin area is the selected study area to deploy the 4ONSE systems. The Deduru Oya is 115km long and covers 2687km2 of catchment area. It carries flash floods during rainy seasons and very low flow during the dry seasons. Hence, the selected case study area is ideal for developing any forecasting application based on the real time environmental data. The project team has already deployed 27 systems in the basin. Temperature, rainfall, atmospheric pressure, solar radiation, relative humidity, soil moisture, wind speed and direction are the parameters measured by the system. In this research we attempt to present the state of the art of open source based environmental monitoring, architecture of the 4ONSE system, motivations behind deployment of the 4ONSE system and finally the lessons for the future research and development.
item: Conference-Full-text
A rule-based lemmatizing approach for sinhala language
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Nandathilaka, M; Ahangama, S; Weerasuriya, GT; Wijesiriwardana, CP
Speech recognition, natural language processing, language translation and deep learning researches are bridging the communication gap between humans as well as between humans and machines. Sinhala is a native language in Sri Lanka which is being used by 19 million people approximately. The growth of Sinhala natural language processing tools is less when compared to European and other Asian Languages. A lemmatizer for Sinhala can be used for the morphological analysis and is an essential module in Sinhala language processing mechanisms. Lemmatizing is a complex process in morphological analyzing where base/root of words are derived. There is not much work published focusing on lemmatizer approaches for Sinhala. This paper presents a rule based lemmatizing approach which can be used to determine the base form of Sinhala words with an accuracy of 77.3%. It differs from similar works because the data used in the research are extracted from social media.
item: Conference-Full-text
An approach of filtering the content of posts in social media
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Kumaresamoorthy, N; Firdhous, MFM; Wijesiriwardana, CP
Social media is playing a major role in relaxing, sharing thought for keep good relationship and academic purpose as well. But, there are some improper or demotivating posts available together, it makes parents to keep their children away from social media and them losing world knowledge. Therefore, keeping good security on content sharing should be considered and filtered out. Newsfeed is one the important portion on Facebook having attractive photos and text. Text of newsfeed posts are filtering out here by following a specific workflow. Text content is taken into the flow initially on feature extraction by non-textual features and replace them with relevant meaning for text processing. Later, the text is optimized by acronyms handling and removing stop words. Then, bad words available text content is separated using two type of dictionaries. By following a specific logic, those sent to classifier for identifying the meaning of the text. Support Vector Machine is used for achieving binary classification and output of this module is content can be viewed to user. For increasing its effective output, similarity among comments with the post is analyzed based on spam detection technique. Advertisements are focused as spam comment here. The system can be plugged with any social media for securing child user from demotivating posts.
item: Conference-Full-text
Integrated corporate network service architecture for bring your own device (byod) policy
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Seneviratne, BLD; Senaratne, SA; Wijesiriwardana, CP
As consumerization of Information Technology occurs, Bring-Your-Own-Device (BYOD) policy is becoming increasingly popular in Corporate Mobility, where organizations allow personal devices to access corporate networks and their services. However, unlike the organizationcontrolled devices, the personal devices pose a threat to the organization’s information resources due to relaxing of access policies to be adopted allowing personal devices to access the networks. This becomes increasingly a challenge for medium scale organizations due to their cost reduction initiatives and inability to implement costly security solutions. Thus, we introduce a network solution for successfully implementing BYOD policy of an organization that allows it to gain complete manageability of access to organization resources and implement policy decisions at both network and individual device level. An organization can create groups of BYOD policies and apply those policies to different containers of mobile devices for controlling and managing mobile devices both online as well as offline. Unlike other solutions, our solution enables applying rules based on individual user, groups, divisions, and device at the network access level by identifying the ownership of individual datagrams; yet can be implemented on an existing operational network without disruptions. The solution architecture includes Mobile Device Management Server architecture and network Access Control Service Server architecture. The complete design of the solution and the future directions are presented.
item: Conference-Full-text
Segmentation of overlapping and touching sinhala handwritten characters
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Walawage, KSA; Ranathunga, L; Wijesiriwardana, CP
Sinhala is the official and national language of Sri Lanka. Seventeen million people of Sri Lanka use Sinhala language to their day to day works. Most of the researches have been done to Sinhala printed character recognition with high accuracy. Nowadays, Sinhala handwritten character recognition is popular research in Sri Lanka. It is not like printed character segmentation; shape of the same type of handwritten character can be changed in different times. Therefore, characters will be overlapped or touched with each other. Handwritten character segmentation is more important to increase the accuracy of the character recognition. Currently there is lack of high accuracy finding to segment overlapping and touching Sinhala handwritten characters. This paper introduced a connected pixel labeling method to segmentation of overlapping characters and peak and valley point identification method to segmentation of touching characters. According to tested result, connected pixel labelling method has 97% accuracy and peak and valley identification method has 72% accuracy.
item: Conference-Full-text
R programming for social network analysis - a review
(Information Technology Research Unit, Faculty of Information Technology, University of Moratuwa, Sri Lanka, 2018) Maddumage, C; Dhanushika, MP; Wijesiriwardana, CP
Social network data have become richer and easier to collect, but that has pushed the borders of the prior art of advanced networks. Therefore, capability of scaling social network is not simply in terms of these advanced networks. However, with the advent of social network, network analysis has always been committed to the development and interaction with social network analysis (SNA). Because of the emergence of continuous SNA in the current state of technological progress, analyzing social network is more conducive to individuals and organizations to make a better future. Even though there are various programming languages which can be used for SNA, this article focuses on use of R programming language due to its huge library of packages. With this background, this article presents a review on R programming in various types of SNA and related issues for best practices.

Browse

Recent Submissions