KICACT 2017

Permanent URI for this collectionhttp://repository.kln.ac.lk/handle/123456789/17369

Browse

Now showing 1 - 20 of 43

V-Synch: Rendering Distance a No-issue with the New Feature of Video Synchronization in Existing Multimedia Platforms.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Tiwari, R.; Shakya, S.
Social media are computer mediated technologies that allow creating and sharing of information idea, career interests and other forms of expression via communities and networks. They introduce substantial and pervasive changes to communication between businesses, organizations, communities and individuals. Various features are being introduced in this field with the objective to make it more attractive to users. “V-Synch” is aimed at introducing features like video and sketch pad synchronization to develop a full- fledged app that also has the current popular features like internet call and chat. We intend to make an android application in which users can always stay connected through multiple platform synchronization (watch the video and use sketch pad in synchronized way in real time) although they are distance apart. All the devices connected to the group can take control of video playback. When any user of that group starts, pauses, or performs specific action on a video, the state of that video is synchronized to all other connected devices in real time. The elements drawn on sketch pad are also shown live in real time to everyone connected to the group. NTP algorithm is used to synchronize all participating devices to within a few milliseconds of Coordinated Universal Time (UTC). The synchronization is correct when both the incoming and outgoing routes between the client and the server have symmetrical nominal delay. V-Synch could be very much beneficial to students for group study, long distance friends to hang out together and Serve a great deal in case of tele-education.
Four Legged Walking Robot with Smart Gravitational Stabilization
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Anthony, A.S.; Pallewatta, A.P.
There are many dangerous jobs which could be safely replaced with an adequately designed robot: bomb disposal; construction in high rise buildings; examination of radioactive environments and combat oriented police/military operations. A machine must then achieve a level of dexterity and reliability greater than that of a human worker. One of the most versatile dynamic robots that can be seen today was made by Boston Dynamics: the quadruped robot named Spot Mini is capable of handling objects, climbing stairs and operating in an office, home or outdoor environment (Bostondynamics.com, 2017). One of the main shortcomings of such robots are their size, cost and inherent need for power. Additionally, a dog inspired gait structure is not optimal for climbing. The aim addressed in this study was to design a robot that would be inconspicuous, capable of maneuvering through small environments and be able to climb inclined surfaces with minimum processing power and cost. To this end, the robot was programmed with an insect inspired gait mechanism for maximum surface area while climbing and a novel ability to maintain the center of gravity by leg movements as shown in figure 1A. Table 1 shows a direct comparison of mobility between the finished robot and an average human being. It would either walk or stabilize once instructed via Bluetooth. The newfangled placement of legs ensured bipod gait during locomotion for faster and efficient motion and monopod gait during the stabilization phase for agility. The desired positions were calculated by the use of inverse kinematics and data from the IMU. The finalized robot was able to successfully walk and proceed through various terrain including grass, sand, small stones and miscellaneous household objects such as books, bags, pencils etc. The auto balancing function worked for as steep an angle as 55°.
Finite Element Analysis of Inflation Tyre Simulation Using Simulia Abaqus.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Fernando, A.; Darshana, D.
Finite Element Analyses (FEA) are conducted in tyre industry either for a safety verification or prediction of the characteristics of virtual tyres. In the design stage of a tyre, it is impossible to determine its manufacturing performance. Therefore, modeling of a virtual tyre plays a significant role when predicting the performance and the characteristics of the tyre. At the end of the virtual simulation if the results do not satisfy the customer requirements, the tyre parameters can be easily changed without manufacturing a prototype sample. This inflation simulation is conducted to determine the outer diameter and the section width of the tyre after pressurizing it under given loading conditions. To obtain the required characteristics of the tyre, the input parameters are adjusted accordingly. It leads to analyze several versions of this virtual tyre simulations. Here, three different versions of virtual tyres are individually analyzed and, the best fitting parameters are determined. The accuracy of the FEA method is estimated by comparing simulation results with that of the prototype dimensions. In the method, three versions of the virtual tyres and the prototype tyres are individually compared to verify the results. As per the estimation, FEA of virtual model simulation shows low dimensional variance (2.38%) compared to that of the actual prototype simulation. Therefore, the results confirm the high accuracy of FEA method in virtual tyre simulation and the importance of implementing it in local industries. It would certainly cause to save precious time, unnecessary cost while increasing the quality of the products.
Introduction of a Four Stage Process of Developing Interactive Multimedia Based E-learning Materials.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Jayantha, R.H.U.; De Pasqual, M.K.; Suraweera, S.A.D.H.N.; Yatigammana, M.R.K.N.; Pathiranage, D.M.; Pallewatta, A.; Wijayarathne, P.G.
Interactive multimedia-based learning materials have been commonly used to facilitate teaching and learning. Technological tools have made the task of creating expression through multimedia more easily available. Invariably this has altered the dynamics of interactions that have traditionally constituted educational ecology of the classroom. Sri Lankan higher education sector has slightly move towards student-centered (collaborative) elearning based around construction to increase equity of access to education, to improve teaching and learning, and to promote students and academic staff in student-centred and activity-based teaching and learning. In designing pedagogically sound interactive multimedia-based e-learning materials, a high premium needs to be placed on leveraging a judicious mix of various presentation modes to cater to user’s differing learning styles and needs. This will ensure that learning is optimized which is essentially student-centred in nature in multimedia rich learning environments. However, as identified by National E-Learning Resource Center (NELRC) at University of Kelaniya, Sri Lanka, most of public higher education institutes largely use face-to-face teaching while e-learning is used as a supplementary tool. There is a lack of understanding of developing technological and pedagogical sound interactive multimedia based e- learning materials which are current problem areas seeking attention. This study used qualitative methodology which made use of qualitative method such as content analysis. This includes three distinct approaches: conventional, directed, and summative. This study used conventional content analysis where coding categories are derived directly from the text data. Based on conventional content analysis of e-learning literature which published in 2010-2016 and retrieved from EBSCO database, the four-stage process i.e. Analysis, Design, Develop,and Delivery has been developed to be used in developing technological and pedagogical sound interactive multimedia based e-learning materials in the Sri Lankan higher education system. After understanding the requirement of developing e-learning materials, the identified process start with the analysis stage which include multiple stages i.e. analyze the needs, cost, content, market, technology, and delivery method and assessment strategies. Design, develop, and delivery stages can be then carried out which also include multiple steps. This process will be useful as a guide for any e-learning centers or any teaching and learning organization for developing interactive multimedia based e-learning materials.
The Impact of a Security Culture in Small and Medium Scale Enterprise (SME) on Enterprise Information Security
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Pathirana, H.P.A.I.; Karunathilaka, J.A.M.A.
An information system is much more than computer hardware; it is the entire set of software, hardware, data, people, procedures, and networks that make possible the use of information resources in the enterprise. In current world, the information is stored in the computerised system in the form of digital data, including sensitive data, which can be extracted as needed. It is much better than maintaining hard copies in traditional manner by using physical storages. The information system security is crucially important for a business with that background. The SME introduces in many forms. Many use the number of employees, capital amount invested, turnover amount, and nature of business. In Sri Lanka, main banks use value of fixed assets as a way to introduce SME, whereas the World Bank uses number of employees as the criteria. Even though enterprises are relatively small and run with a limited budget, SMEs can now target national and international market segments, enabled by the Internet. Therefore, this complicated the business process at SMEs. The computer security represents confidentiality, integrity and availability (CIA) from the mainframe-computing era. The rise of the Internet and complex computer systems means that data is now decentralized. As such, the security measures now must extend form the CIA domain to cover additional areas, depicted in the McCumber Cube in three dimensions. This challenges SME’s to assure information security with a limited operating budget, and there are two approaches presented by the ‘Sphere of Protection’, focusing on both technology and people aspects. The technological aspect is expensive, whereas the people aspect is cost effective by introducing security culture. The policy implementation is the better tool for security culture by considering business in process level emphasizing laws to acknowledge people on the importance of assuring secure environment, and education and training are important to share the knowledge among employee. This paper explores the need for effective people based security measures for better security culture, before the implementation of technological controls is considered for SMEs.
A Simple Machine Learning Approach for Identifying Promotional Short Message Service (SMS) Messages.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Dias, D.S.; Dias, N.G.J.
Mobile phones play an integral part in the modern lives of humans. Short Message Services (SMS) Messages have become a popular mode for simple communication. Its’ simplicity, costeffectiveness and large audience has attracted the attention of advertising industry to send targeted promotional messages to mobile phones. In Sri Lanka, a survey conducted in Colombo, yielded that 3 out of 5 SMS messages received our promotional messages. Even though extensive research has been carried out in detecting junk SMS messages, the amount of research conducted on filtering promotional SMS messages is rare. The purpose of this research is to evaluate the success and accuracy of utilizing a simple machine learning algorithm to identify promotional SMS messages. Here, we have used a feed-forward neural network based on a statistical model, which was trained with a training data set consisting of promotional as well as non-promotional messages. Each test message was broken down in to individual words and filtered through by cleaning to form keywords which will have consist of a weight and probability value. With each message that is used to train, these values will be updated according to whether it is a promotional or a non-promotional message. When a message is tested through this neural network, the words of the message will be matched against the keyword’s weight and probability, which will then calculate a resultant probability. By setting a par-value, we can classify the test as a promotional or a non-promotional message. The proposed model yielded a 100% accuracy when tested within the given test data set. In order to get successful results for broader test data sets, the model has to be trained comprehensively with proper amount of promotional and non-promotional messages. Optionally, the results obtained from the feed forward neural network for incoming messages, can then be fed back in to the feed forward neural network for further training. As future work, we intend to take this solution to an android-based mobile application that extracts promotional messages from the incoming SMS messages as well as from a server, and display them to the user based on his preferences.
Facilitating an E-Learning Platform Beyond the Lectures: Digital Natives Become Active Learners.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Weerakoon, A.D.
The traditional lecturing can't inspire the digital natives towards the engagement in active learning to succeed them in the university environment and beyond in the real world context. In 2004, Romiszowski declared that e-learning presents an entirely new learning environment for students, thus requiring a different skill set to be successful. In 2008, Markus stated that e-learning is a learning process created by interaction with digitally delivered content, network-based services and tutoring support. E-learning is also called web-based learning, online learning, distributed learning, computer assisted learning, or internet based learning. This study was focused to explore the impact of a poster exhibition project on the active learning of digital natives by providing an e-learning environment. This study was carried out with Level 2 Polymer Engineering Technology students and four consecutive annual poster exhibitions has been conducted with four different batches. Each poster exhibition project was a one-month project. The students were grouped into 12 teams of 2 students in each group and each group had to prepare one poster after finalizing a theme for the poster exhibition project and the topics for the individual posters. The theme and the topics were selected to cover more than the 75% of the syllabus content of DPT 207 Polymeric Materials subject. In preparation of the posters, each group had to write a report in prior to create the rough skeletons for the poster by referring relevant articles including journal articles through the internet and each group was asked to email that report to the researcher before the given deadline. Through the constructive feedback the students had to modify the rough skeletons several times and finally came up with amazing posters. At the end of the poster exhibition project the students were given a questionnaire with both open-ended and close-ended questions. Descriptive statistical results reveal the facilitation of e-learning helps the students to learn actively, motivationally and to enhance self-monitored learning along with the collaborative learning. By enabling learners to be more active participants, a well-designed-e-learning experience can motivate them to become more engaged with the subject content and further develop them as lifelong learners.
Applying Intelligent Speed Adaptation to a Road Safety Mobile Application –DriverSafeMode
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Perera, W.S.C.; Dias, N.G.J.
During the last decades, Sri Lanka has experienced a highly accelerated growth level of motorized transportation with the rapid urbanization due to the economic development. However, the increasing motorization has also placed a significant burden on people’s health in the form of uncontrollable growth rate of road accidents and fatalities. We have focused on excess speed and mobile distraction which are two major factors that have caused majority of road accidents. Exceeding the speed limit, which is enforced under the traffic law, increases both the risk of a road crash as well as the severity of the injuries by reducing the ability to judge the forthcoming events. Use of mobile phones distracts a driver in the means of visual, physical and cognitive. These factors are largely preventable but are unlikely; due to the lack of adequate mechanisms in existing road safety plans in Sri Lanka. Especially in rural areas, roads are poorly maintained which has led to faded, hidden, foliage obscured speed limit signs and absence of appropriate signs at vulnerable locations (schools, hospitals). Existing plans also lack alert systems to avoid drivers from using phones while driving. Proposed application uses Advisory Intelligent Speed Adaptation (ISA) to ensure drivers' compliance with legally enforced speed limits by informing the driver on vehicle speed along with speed limits and giving feedback. There exist many ISA systems deployed using various methods such as GPS, Transponders, compasses, speed sensors and map matching, based on native traffic infrastructures of other countries. Google Fused location provider API web service was used combined with GPS sensor of the smartphone to obtain continuous geo location points (latitude, longitude). Distance between two location points was calculated using Haversine Algorithm. Using the distance and time spent between two location updates, vehicle speed was calculated. Google Maps Geocoding API was used to obtain the type of road on which the driver is driving. Accepted speed limits were stored in a cloud hosted database according to each road type and vehicle type. Application establishes a connection to the database to gain the accepted speed limit whenever a new road type is detected. It compares real-time speed Vs speed limit and initiate audio and visual alerts when the vehicle speed exceeds the limit. Google Places API was used to identify schools and hospitals within 100m and informs the driver using audio and visual alerts. Application uses in-built GSM service to reject incoming calls and in-built notification service to mute distracting notifications. A test trial was carried out to evaluate the accuracy of speed detection. Mean speed of the test vehicle speedometer was 14.4122kmph (Standard Deviation=14.85891) and that of the application was 13.7488kmph (Standard Deviation=14.31279). An independent-sample t-test proved that the speed values of the test vehicle and the application are not significantly different at 5% level of significance. User experiences of 30 randomly selected test drivers were evaluated. 80% of lightmotor vehicle test drivers had stated that the application is very effective. 10% of the heavy-motor vehicle drivers and 20% of tricycle test drivers had found it difficult to perceive the audio alerts due to the noisy surrounding. Evaluations prove that the usage of the proposed system can impose a direct and positive effect on the road safety of Sri Lanka as expected.
Stock Market Analysis and Prediction.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Shakya, A.; Pokhrel, A.; Bhattarai, A.; Sitikhu, P.; Shakya, S.
Stock price and stock index price forecasting system, used by investors and financial managers to describe the market and compare the return on specific investments, has been a topic of research for very long now. When in the stock market, there are more buyers than there are sellers, the price must adapt or no trades are made. This tends to drive the price upwards, increasing the market quotation at which investors can sell their shares, enticing investors who had previously not been interested in selling and vice versa. These demands and supplies are ever changing, resulting in highly-fluctuating, non-linear stock prices which poses a threat against the credibility of those prediction systems which only view the market from one perspective. For a reliable system, it is therefore important to explore the market on multiple grounds, basically through Technical, Fundamental and News Analysis. Under Technical Analysis, SMA (Simple Moving Average) is used as a preliminary data smoothing technique, which helps reduce the fluctuations substantially. Artificial Neural Networks (ANNs) is then employed to analyze the nonlinear relationships between the stock closed price and various technical indexes, and to capture the knowledge of trading signals that are hidden in historical data. Features like traded share, traded volume, opening price, closing price, high price and low price are fed as an input parameter in Neural Network. Backpropagation algorithm is then implemented to train the given Network model. The neural network layers and neuron numbers in hidden layers are then tuned by training and validating the data set iteratively. The fundamental analysis involves thorough study of financial statements of companies, also known as quantitative analysis. This involves looking at assets, liabilities, revenue, expenses and all other financial aspects of a company. It gives insight on the company's future performance. The results moreover reflect the company's success or failure over the long term than immediate future. Hence, unlike technical analysis, it helps predicting stock price on a long run. In news analysis, we focus on understanding the news sentiment and its affects which may cause the investors to either buy or sell the shares based on positivity or negativity of the news. The news analysis problem can be mapped into similarity based classification. A set of vectors are created from analysis of historical news, where each component of a vector represents the features in data set. The required labeling are done based on historical rise/fall of stock prices. The increase or decrease of the trend is then predicted based on similarity measures. In short, news analysis predicts the price of share of the following day by comparing the most recent news with past news using Knearest neighbor algorithm. Thus, through the circumstantial application of the above-mentioned analysis, the paper proposes to predict the stock market in a more generalized manner.
Design and Development of a Dashboard for a Real-Time Anomaly Detection System.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Korala, H.C.; Weerasooriya, G.N.R.; Udantha, M.; Dias, G.
Web logs contain a wealth of undiscovered information on user activities and if analyzed in a proper way they can be utilized for many purposes. Identifying malicious attacks and having a daily summary on user activities are some valuable information that can be extracted from these log files. At present, many tools and algorithms have been developed to extract information from these log files but on most occasions, they have failed to present this information to the user to make decisions in real-time. This paper presents a novel approach taken to design and develop a dashboard for a real-time anomaly detection system with the use of some open source tools to process complex events in real-time, batch process stored data using big data tools and dashboard development techniques. The system accepts web log files as the input; first they are cleaned by a preprocessing unit and then published to WSO2’s complex event processor as events to identify and filter out special patterns and summarised by using a set of user specified rules. If an anomaly is detected, an alert or warning will be displayed on the widget based dashboard in real time. Furthermore, each and every event stream that comes to the CEP will be forwarded to WSO2’s Data Analytic Server via 'Thrift' protocol. That data will be saved in a Cassandra big data database for further batch processing which is used for drill down purposes. A widget based Dashboard has been developed with the use of modern dashboard concepts and web technologies to display information such as daily summary, possible security breaches in an interactive way allowing system administrators to make operational decisions then and there based on the information provided. Moreover, users can drill down and analyze the historical security breach information and also can customize the dashboard according to their preference. The evaluation techniques used fall under the criteria of evaluation against well-established standards and evaluation by external expert review. Evaluation for security standards has done against the security standard set by the PCI security standards council and evaluation for dashboard has been carried out against the dashboard standards defined by Oracle which describes about the best practices in developing an effective dashboard. Evaluation by external expert review was done in line with the people who have prior experience in dealing with a dashboard in different contexts. Ten expert evaluators from different expertise areas (System Administrators, UX engineers and QA engineers) have been used for this evaluation and a score based model was used to determine how efficient this dashboard is to view and drill information. Based on the results yielded from the evaluation, it is identified that the dashboard meets with the international standards of dashboard designs, well established security standards in dashboard design as well as provides the best user experience for users in different functional areas.
Challenges in Implementing ERP Systems in Small Medium Manufacturing Companies in Sri Lanka
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Yasotha, R.; Ramramanan, L.
There are numerous information systems available in the market to be selected for implementation in manufacturing organizations. When many information systems manually intergraded for management reporting for a company, there are high risks for accuracy of information. ERP is one of the information systems with inbuilt capacity to integrate many parts of the functional areas that provides meaningful information to the management. This paper describes the experiences on how a small medium size growing roof manufacturing company in Sri Lanka problem and then overcome in implementing ERP system. Small medium size manufacturing companies in Sri Lanka do not normally have electronic information system in all part of business process, whereas some processes such as production process operates outside the information system. Therefore, it is very important to predefine what level of integration to be done, who are the related parties to be consulted and what level of management information is required. The success of ERP implementation is partially depending on the selection of suitable ERP system compatible with company business process and the capability of implementation partner to map those standardized business processes into ERP by conducting BPR. This manufacturing company has many automated manufacturing plants with Programmable Logic Controllers (PLC) versions from year 1960 to 2013. When these PLCs try to integrate into ERP system, there are so many problems faced by the company that leads up to modification of plant. Finally, company decided to implement ERP by postponing the PLC integration. Well tested bugs free less customized SAP B1 system has been implemented to the company by monitoring progress by several log books. The big bang approach has been followed to implement the SAP B1 system with short term parallel run of legacy system. More importantly, top management support and motivation on change management has fuelled up the success of the SAP B1 implementation. This paper reveals the experience gained during the planning to implementation stages of SAP B1 that may occur in small medium manufacturing companies in Sri Lanka.
Detecting and Classifying Vehicles in Video Streams of Homogeneous and Heterogeneous Traffic Environments Using Gaussian Mixture Model.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Jayathilake, M.V.M.; Jayalal, S.G.V.S.; Rajapakse, R.A.C.P.
Traffic and transportation play an important part in modern national economics. Efficient use of transportation infrastructure leads to huge economic benefits. Traffic can be classified into two main categories as homogeneous traffic and heterogeneous traffic. In transportation engineering, sufficient, reliable, and diverse traffic data is necessary for effective planning, operations, research, and professional practice. Even though, Intelligent Transport System are used to find answers for that issue still it is not yet fully successful. Many technologies have been developed to collect different types of traffic data. Traditional data collection technologies have several drawbacks. On the other hand, video based traffic analyzing has become popular. Computer vision techniques are used for detecting and classifying data in traffic videos. Those technologies are highly beneficial as it can give us more information about the parameters, easy to install and maintain and has got wide-range operation. In Computer vision, vehicle detection process has two main steps as Hypothesis Generation (HG) and Hypothesis Verification (HV). Background Subtraction is a popular method used in HG. There are several algorithms used in Background Subtraction and Gaussian Mixture Model is one of them. These methods are used in homogenous traffic situations. The objective of this study is to detect and classify vehicles from a homogenous and heterogeneous traffic video stream using Gaussian Mixture model. This study was conducted using an experimental method. Several set of road traffic videos were collected. One is collected at off peak time; i.e. 9.00am to 10.00am. At that time behavior of the traffic is similar to homogenous traffic environment. The other set of videos is collected from 7.00am to 8.30am. At that time, road traffic has no order and the traffic density is high. It is similar to heterogeneous traffic environment. After Gray Scaling and Noise reduction, the videos were submitted to algorithm based on Gaussian Mixture Model. The algorithm was implemented using Math Lab software. Vehicles are classified as large, medium and small. Manual observation results and experiment results were compared. Accurate results were observed from homogenous traffic conditions. But results in heterogeneous traffic conditions is less accurate. The Gaussian Mixture Model can be used to detect vehicles in homogenous traffic conditions successfully, but it is needed to be improved in heterogeneous traffic conditions.
Factors Influencing the Students’ Intention to Adopt E-learning Special Reference to Eastern University, Sri Lanka.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Nirushan, K.
E-Learning is becoming an important part of learning process. With the evolvement of Information Technology, the “Teacher Centered” traditional learning methodology has started to change to “Learner Centered” methodology. As per this change in learning process, the use of technology plays an important role to enable students to engage fully in their program of study. Moreover, elearning process makes the students very easy to engage with their academic activities. In most of the developed countries, “Distance learning” became huge popular with the use of e-learning process. In Sri Lanka, also most of the higher institutions are trying to provide e-learning facilities to their students in order to utilize the advancement of modern technologies. However, it is necessary to identify the influencing factors regards to e-learning process to fuel the utilization of this emerging technologies such as Virtual Classroom, Learning Management System (LMS). This study examines the influencing factors on students’ intention to adopt e-learning as a tool of learning. Therefor 210 students were randomly selected from Eastern University, Sri Lanka and data were collected through a structured questionnaire. Correlation and Multiple regression analysis were done based on the Technology Acceptance Model (TAM). More than this model a variable called “Prior Knowledge on ITC” was added and analysis was run. Correlation denotes that Perceived ease of use has significant medium positive relationship with intention to adoption of e-learning where r=0.483, p=0.000<0.01. Perceived usefulness and prior knowledge has significant positive strong relationship with intention to adoption of e-learning where r=0.773, p=0.000<0.01 and r=0.863, p=0.000<0.01 respectively. However, multiple regression analysis reveals that “Prior knowledge in ICT” is the most influencing factor on intention of adoption to the e-learning activities. Chi-square test confirms that there is a difference between two gender group in intention of adoption to e-learning activities and crosstabulation analysis shows that boys are more intent to adopt e-learning activities than girls.
De-Identification for Privacy Protection in Audio Contents
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Induruwa, K.G.; Pallewatta, A.P.
Among different forms of audio data or information, the author wishes to limit the scope of this research to privacy protection in voice contents of speakers, because voice generally conveys intelligence such as gender, emotion and it differs from speaker to speaker. De-identification of voice may bring numerous advantages, such as preserving the privacy of speakers during communication, maintaining confidentiality of inquirers who conduct critical investigations and improve the clarity of voice signals used in airport/aviation communication by standardizing the voices of Pilots and Air Traffic Controllers. Though advanced voice encryption methods are available to deteriorate the intelligence of speech, they do not directly address the issues of speaker de-identification. This research project aims at de-identification of voice signals while preserving the intelligence of the speech during communication. Designed GUI for mono LPC spectrums of original and de-identified voice signals In this project, the de-identification process was done at three stages, where the last two processes are irreversible. First, in the frequency normalization stage, pitch of the original signal is changed and slightly de-identified the voice in frequency domain. Then 12 LPC (Linear Predictive Coding) co-efficient values of the subject-person’s original voice signal is subtracted from the 12 coefficient values of the reference sample voice signal. As a result, features are slightly moderated by the second stage. In the third stage the features are destroyed again by shuffling LPC coefficients randomly within three categories. Therefore, this whole process is expected to preserve a higher level of privacy. Based on the test carried out by using 15 samples of male and 15 samples of female voice produced a degree of 10% and 20% de-identification, which could be accepted as a very satisfactory result.
Automated Characters Recognition and Family Relationship Extraction.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Bajracharya, A.; Shrestha, S.; Upadhyaya, S.; Shrawan, B.K.; Shakya, S.
“Automated characters recognition and family relationship extraction” is an application of Natural Language Processing to identify characters from the story and determine the family relationship among them. This application is the use of specialized computer programs to identify entities, classify them and extract characters from them and determine relationship between them. This paper follows basic steps of NLP i.e. Tokenization, POS tagging, sentence parsing followed by the pronoun resolution implementing various algorithms and finally extracting entities and relations among them. Heretofore, we have successfully resolved pronoun from simple sentences by resolving Noun Phrase using the recursive algorithm for tree generation and hence extracting relation between the Noun Phrase (NP). Basic approach towards this project is to do Tokenization and POS tagging first. Then, sentence which is recursive composition of Noun phrase, verb phrase and prepositional phrase is parsed and recursive tree is generated. Then tree is traversed to determine the noun phrase which is replaced by the entity object of that particular noun phrase. Pronoun resolution is the essence of NLP and is of different type. Here, Co reference resolution has been used. After resolving the entire pronoun, then finally relationship is extracted from the story by comparing the relation ID of each Entity. Given the simple story, entities are being extracted and relationship is also determined. Understanding the approach of NLP and implementing them to showcase its use is the main theme of this project which is being done with as accurate result as possible. This paper can act as a base.
Artificial Neural Network based Emotions Recognition System for Tamil Speech.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Paranthaman, D.; Thirukumaran, S.
Emotion has become the important part in communication between human and machine, so the detection of emotions has become important part in pattern recognition through Artificial Neural Network (ANN). Human's emotions can be detected based on the physiological measurements, facial expressions and speech. Since human shows different expressions for a particular emotion when they are speaking therefore the emotions can be quantified. The English speech dataset is provided with descriptions of each emotional context available in Emotional Prosody Speech and Transcripts in the Linguistic Data Consortium (LDC). The main objective of this project describes the ANN based approach for Tamil speech emotions recognition by analyzing four basic emotions sad, angry, happy and neutral using the mid-term features. Tamil speeches are recorded with four emotions by males and females using the software “Cubase”. The time duration is set to three seconds with the sampling frequency of 44.1 kHz as it is the logical and default choice for most digital audio material. For the simulations, these recorded speech samples are categorized into different datasets and 40 samples are included in each dataset. Preprocessing includes sampling, normalization and segmentation and is performed on the speech signals. In the sampling process the analog signals are converted into digital signals then each speech sentence is normalized to ensure that all the sentences are in the same volume range. Next, the signals are separated into frames in the segmentation process. Then, the mid-term features such as speech rate, energy, pitch and Mel Frequency Cepstral Coefficients (MFCC) are extracted from the speech signals. Mean and Variance values are calculated from the extracted features. To create the classifier for the emotions, the above statistical results as an input matrix with their related emotions-target matrix are fed to train, validate and test. The neural network back propagation algorithm is executed by the classifier to recognize completely new samples of Tamil speech datasets. Each of the datasets consists of different combinations of speech sentences with different emotions. Then, the new speech samples are assigned to identify the recognition rate of the speech emotions using the confusion matrix. In conclusion, the selected mid-term features of Tamil speech signals classify the four emotions with the overall accuracy of 83.45%. Thus, the mid-term features selected are proven to be the good representations of emotions for Tamil speech signals and correctly recognize the Tamil speech emotions using ANN. The input gathered by a group of experienced drama artists who have the voice with the good emotional feelings would help to increase the accuracy of the dataset.
Smart Home Automation Voice Controller.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Perera, P.V.S.P.; Weerasinghe, K.G.H.D.
Each day we are aiming for a smart living condition and make our lives more convenient and fast. The traditional wired electrical device controlling switch is an old concept now. “Voice operated device controlling” utilizes human voice commands to control electrical appliances. This research aims to design and implement a cost effective, portable, user-friendly, secure and simpler Home automation voice controller that can be operated by using Android smart phone. It also reduces the energy usage in the residential sector. This system is also designed to assist and provide support in order to fulfill the needs of elderly and disabled in the home. This research describes the way of remote controlling and monitoring electrical household appliances using Android Smart Phone Bluetooth features and wireless Bluetooth technology module depending user voice commands. The proposed system has two main components, namely voice recognition system and clicking mode facility. When automating a home load not available in the visible range, fault identification system in this design helps the user to ensure that their home appliances had gone exactly ON or OFF. The app was designed by allowing the user to add or edit the appliances. The user had the freedom to add appliances names to this app. User can select either voice mode or clicking mode. Even he/she can check the current status. Changing the language is also available in this app. As an example device name is Fan. The user has to say “Fan” to switch ON. If user wants to switch off, again, has to say “Fan”. Google voice recognition with its voice recognition and voice command features has been used to determine the voice of the user. From the commands received from an android device, the electrical appliances’ current status can be controlled. Android Phone will convert voice into a string of data using Google voice recognition feature. This string of data will be sent to Bluetooth module and then to Arduino UNO. After that, Arduino decodes and process it. The Figure 1 expresses the system architecture of the entire system. Arduino UNO is very popular, cheap product and very easy to use. Bluetooth module, relays are interfaced to the Microcontroller. The data received by the Bluetooth module from an Android smart phone is fed as input to the controller. The controller acts accordingly on the relays of the electrical appliances. The electrical appliances in the research can be made to switch on or off using the Android phone. The application shows the status of switch whether on or off. In achieving the task, the controller is loaded with a program written using Arduino language. This system facilitates features such as automation, multi-functionality, adaptability, interactivity and efficiency for home appliances controlling. As future enhancements, hope to design input voice commands in different language and hope to design smart watch with hand gestures to control in a more user friendly.
Machine Learning Dashboard for Aviation Fuel Optimization.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Samarasinghe, R.M.N.S.; Dias, N.G.J.
The aviation industry is the one of the fastest-growing travel industry in the world. This industry is growing 7% per year and is giving its facilities for more than 1.5 billion passengers. The International Air Transport Association (IATA) indicates that this number will pass in the next 20 years by 7.3 billion of passengers. Due to this large growing passenger count, airplane manufacturing companies such as Boeing & Airbus are making more efficient planes to handle this amount. Aviation fuel is the biggest cost in air transport. IATA (The International Air Transport Association) figures show that everyone dollar increase in the cost of oil per barrel increases the airline industry's costs by about $1 billion. So that airline companies do their best to optimize the fuel usage managing many types of maintenance, weight flowing management to reduce the plane taxi fuel. Airplane manufacturing companies are also gearing up to make more fuel-efficient planes. This research project built finding suitable variables and providing a solution to overcome the high fuel usage by using a neural network model to predict the fuel usage, CO2 emission dashboard to get necessary steps to reduce CO2. Finding the suitable variables are the most challenging part in this research. To find them, correlation coefficient method was used. Before using this method need to normalize the dataset using the statistical normalization method after that used this method to find the linear combinations of the fuel usage & other dependent variables. If the value is next to -1 then it gives a perfect negative relation or if +1 then it is a perfect positive relation. For this analysis, the best fit regression model was created based on the variables Actual passenger count, Flight wing size, Flight length, Flight height, Distance between airports, zero fueling weight identified are those variables. For a prediction model, it is more practical to use simple model than a complex model. Before developing this model, data need to be clean (without empty data sets) and eliminate the outlier data from the data set after the normalization process which was done by using the statistical quartile method. For this model 2 types of training, functions were used to create the models ‘Bayesian regularization back-propagation’ and ‘scaled conjugate gradient back-propagation’. ‘Bayesian regularization’ method is the best training to train noisy data sets. After training these 5 layers (4-hidden layer) 5-10-5-10 hidden neuron model, then it was selected as the minimal error rate. There were 26, 834 data points & 70% were used to train this model and the rest 30% was used for testing. For this research, there are lots of future works could be done adding weather data, giving a recommendation in flight scheduling process.
Foreign Exchange Rate Prediction using Artificial Neural Network and Sentiment Analysis
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Shrestha, S.; Baral, S.; Subedi, S.; Ranjit, S.; Shakya, S.
Foreign currency exchange plays an important role for currency trading in the financial market. Modern approach to the foreign currency exchange market requires support from the computer algorithms to manage huge volume of transactions. There occurs problems like trading without a plan, failing to adapt to the market, having unrealistic expectation and many more. Due to these problems, predictions are to be done. This paper investigates on prediction of foreign exchange market using neural network and sentiment analysis. There are many algorithms for performing prediction but different algorithms have different accuracy. One of the best method with high accuracy is given by Artificial Neural Networks (ANN). Neural network parameters include number of hidden layers, number of neurons, use of bias neurons, activation functions and training methods. Input nodes are price of gold, crude oil, Nasdaq index, yesterday’s price. Our model contains 4 input node, 1 hidden layer and 7 hidden nodes. At first, pre-processing is done and inputs are fed to the neural network. By using backpropagation algorithm, training is done and then testing is performed. Mean absolute percentage error is found to be 0.39%. The price movement is also directly related to market sentiment. We aim to employ a statistical technique to the opinion of different traders and finding the overall sentiment. Sentiments are taken from tweets and then filtering the tweets are performed. After that, features are extracted and by using Naïve Bayes algorithm, the results are classified as positive or negative.
Conflict Categorization of ERP Implementations in Asia Pacific Region.
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Herath, H.M.P.S.; Rajakaruna, J.P.
An Enterprise Resource Planning (ERP) system is an integrated software system, typically offered by a vendor as a package that supports the seamless integration of all the information flowing through Business Processes, Business Intelligence, Business Integrations, Collaborations, etc. This research is intended to discuss on complications in ERP implementation in Asia Pacific (APAC) region with the client, vendor, implementer, consultant and project management perspectives. The objective of this research-in-progress paper is to develop a clear visibility of categories of conflicts in ERP projects in multicultural environments. Categorization of ERP project implementation related conflicts would provide better preparation for a successful project implementation and delivery. This is the first attempt for the journey to consolidate the literature on the conflicts associated with ERP projects. Also seeking for uplift the understanding of conflict and managing the same effectively in APAC region. In this case our research question is “Can we categorize ERP project related conflicts?” and if so, “What are the categories of conflicts in relation to ERP implementation in APAC region?” Alsulami (2013) on his “Consolidating Understanding of ERP Conflicts : A dialectic Perspective, Computer Science and Information Systems Faculty, Umm Al-Qura University” categorised ERP projects conflict related to Australian experience into two; such as “Technical related and Process related”. However, thirteen business cases in Sri Lanka, India and Malaysia show us conflicts can be categorised as “People related, Technology related & Methodology related”. These findings can be effectively used by ERP Implementers, Vendors, Consultants, Project Managers and Researchers in their respective projects.