ICACT 2019

Permanent URI for this collectionhttp://repository.kln.ac.lk/handle/123456789/20316

Browse

Now showing 1 - 20 of 31

The Study on the Factors Influencing to Customer Adaption of E-Banking in Sri Lankan Banks
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Perera, P.A.H.L.; Dissanayake, D.H.S.W.; Deshika, N.P.T.
This study attempts to examine the factors affecting to customer adoption of E-banking in Sri Lankan Banks. E-banking provides many services for their customers. Convenience, speed, efficiency, effectiveness are advantages of E-banking on the customer’s hand. Bank considers it as it holding customer and reduce paper works. Through this observation, researcher found factors of people adopt to E- banking and after that Banks can address to those factors and promote it. So they will be able to speed their E-banking market. On the other hand, Customer will get more benefits of E- banking. This study is based on positivistic paradigm hence deduction method is used as reasoning approach and used quantitative techniques. The data for this study is used primary and secondary data to analyze the database and give an opinion. Primary data was collected from customers from five registered commercial banks by using a questionnaire which is a type of Likert scale form. Secondary data was collected by the annual reports (2017) of commercial banks. As the sample of this research, the researcher selected 163 E-banking customers of five registered commercial banks. The researcher has used usage of E-banking as dependent variable and attitude, subject norms, perceived behavior as independent variables. In this study, researcher used descriptive statistics for the determinants of customer adoption of E-banking. The study is used correlation analysis to investigate any relationship between attitude, subject norms, and perceived behavior with the Usage of E-banking. According to the Correlation Analysis, the values of 0.865, 0.689, and 0.688 at 0.000 levels respectively depict the positive relationship between usages of E-banking. The study is used Multiple Regression Analysis to assess the impact of factors influencing to customer adoption of E-banking. According to Regression Analysis, the study reveals that the attitude, subject norms and perceived behavior impact positively to usage of E-banking. This research is significant to all the banks to take their decisions to satisfy their customers. Moreover, by proceeding those decisions banks can achieve their financial targets.
Use of Processed Tea Waste Powder and Fiber in Improving the Properties of Rice Husk Ash Filled Compressed Stabilized Earth Blocks
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Madhushani, K.L.; Yatawara, M.D.M.D.W.M.M.K.
Rice husk ash (RHA) and processed tea waste (PTW) are major agricultural wastes. A recent study has shown that 7.5% of soil can be replaced with RHA in Compressed Stabilized Earth Blocks (CSEBs) due to its pozzolanic properties. Since PTW shows good pore-forming ability, the present study was planned to determine whether there is a potential to improve properties of previously upgraded CSEBs by replacing soil with PTW powder or fiber. The mixing percentage of soil, cement and RHA was 86.25: 6.25: 7.50 in the previously upgraded block. Five types of CSEBs of 300 mm x 150 mm x 100 mm (L x W x H)) were manufactured by replacing 0% (Control- BC), 3 %(B3), 5% (B5), 7% (B7) of soil by PTW powder and 3% (BF) of soil by PTW fiber. The suitability of properties of raw materials were tested prior to manufacture CSEBs. Mechanical properties of manufactured CSEBs were tested. Data were subjected to One-way ANOVA followed by Tukey’s pair wise comparison in MINITAB 14. The values were compared with SLS 1382: 2009 and British standards. The dry and bulk densities and weight reductions of PTW incorporated blocks (B3, B5, B7 and BF) showed significantly higher reductions (p <0.5) than that of Controls. B3 showed the highest compressive strength (3.8 Nmm-2) except Controls. In addition, B3 also showed the lowest surface erosion (pitting depth = 0 mm and pitting rate = 0 mm min-1) and the highest durability (Slake durability Index = 90). In addition, B3 showed the lowest loss on ignition (12.5%) and the percentage weight reduction (6%) except controls. According to SL standard 1382 part 1: 2009, only blocks BC and B3 were suitable for construction of walls (Grade 3). In addition, B3 was also suitable for external walls compared to control blocks. When compared with the British standards, only BC was suitable for load bearing walls for two storey houses. Considering all aspects, 3% of PTW powder incorporated CSEBs with 7.5% of RHA can be recommended for single story buildings and for external use in places where PTW is highly abundant.
Investigation of Lead Concentration in Road Dust Samples in Kiribathgoda Area, Sri Lanka
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Silva, K.D.M.; Premaratne, W.A.P.J.
Heavy metal contamination has become one of the major problems in metropolitan cities all around the globe. Kiribathgoda area in Sri Lanka is one such hot spot. Anthropogenic activities have resulted in the increment of heavy metal levels in the earth’s crust. Analysis of outdoor dust is a useful technique to determine the heavy metal content in an urban area and thereby predict the extent of air pollution. This could open opportunities to relate the threat for human health by such toxic heavy metals in an unhealthy environment. This investigation was carried out by selecting a section from the main road of Kandy-Colombo in Kelaniya area. Ten sampling sites were selected and samples were collected as triplicates for three consecutive months. Their pH level and organic matter content were tested. Also, concentration level of the heavy metal Pb was determined using atomic absorption spectroscopy. The preliminary factors that favor the persistence of heavy metals in the environment were investigated and analyzed. It was observed that slightly acidic (pH 5.89±0.41) dust favor Pb deposition. The Pb content was found to be fluctuating around 22.01-84.52 mg kg-1. A good positive correlation (Correlation Coefficient 0.878) was observed in between Pb- Organic matter. It was evident that Pb exists in the environment for a very long time but their escape from nature is very slow. Hence it is necessary to study and understand the health risks associated with heavy metal toxicity on future studies developing from the findings of this research that will benefit the mankind.
Sorptive Removal of Lead (II) from Aqueous Solution using Value Added Tea-Waste Biochar Produced Under DifferentTemperatures
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Kavinda, K.D.T.; Peiris, C.; Gunatilake, S.R.
The removal of lead from aquatic systems using biochar (BC) derived from tea-waste was evaluated. The customized in-house method of BC production was incorporated slow pyrolysis at 300 ℃ (300BC), 500 ℃ (500BC) and700 ℃ (700BC). The different BC types were subjected to a nitric acid modification and magnetization. Results showed reduced adsorption capacities for nitric modified BC. Batch sorption experiments were conducted to evaluate the effect of pH, equilibrium time, associated kinetic models and the thermodynamic basis of lead uptake. For both Non-Modified Biochar (NBC) and Magnetized Biochar (MBC), an acceptable fit for the pseudo second order kinetic model with regression coefficients greater than 0.998 justified a chemisorption process. The dominant mechanism for 700BC can be considered as pore filling together with π electron sharing between the graphene rings and lead whereas sorption on 300BC was governed by electrostatic interactions. Adsorption isotherms modeled were Langmuir, Freundlich, Sips, Redlich- Peterson and Toth, out of which the results were seen to best fit Langmuir and Sips models. A maximum Langmuir capacity of 57.80 mg/g and 48.61 mg/ for 700NBC and 700MBC were obtained respectively. Positive enthalpies and free energies indicated a nonspontaneous and exothermic sorption. Magnetic modification decreased sorption capacities by 15.86 % but led to the easy removal of biochar after the sorption.
An Application of Artificial Neural Networks to Predict the Milk Yield of a Typical Dairy Farm in Sri Lanka
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Hewage, S.S.; Chandrasekara, N.V.
It is quite interesting that milk and dairy products play an important role in a healthy, balanced diet thus contributing to certain indispensable nutritional benefits. Hence, the need for dairy is crucial, which means dairy farms provide a vital necessity to the people in both rural and other areas across the country. Therefore, accurate forecast of milk yield is important for dairy farmers to utilize and optimize their production process. The present study is aimed at using Artificial Neural Networks (ANN) for predicting the milk yield of a dairy farm by considering the potential factors that affect the milk production. Further, it is important to note that this dairy farm has kept records of the daily milk yield, the amount of food given to cows, and weather condition. Data from January 2016 to June 2018 were used for the study. In this regard, a feedforward neural network (FFNN), non-linear auto regressive neural network (NAR), and a non-linear auto regressive exogenous neural network (NARX) were fitted. Analysis was done using Matlab software and all three implemented models took around 30 seconds for execution. While all the three models exhibited quite strong model performances, the NARX model exhibited prominently outstanding results. The best forecasting performance was shown by the NARX neural network which contained one hidden layer with five neurons having saturating linear transfer function. Normalized Mean Squared Error (NMSE) was 0.0247 for the overall model while the Mean Absolute Error (MAE) value was 6.6245.
3D Visualization of Human EEG Signals
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Pradeep, H.B.A.C.; Meegama, R.G.N.; Kalinga, S.
The brain is the most important and the most complex human organ that is responsible for all the functions that we do in our routine life. Moreover, the brain consists of millions of neurons that utilize electro-chemical signals to transmit information to other parts of the body. Whenever a neuron triggers an electrical impulse to another neuron, it generates electricity, referred to as an EEG wave that can be measured by a sensitive device. Using such brain patterns, it is possible to identify normal day-to-day human behavior. The brain commences its work before birth and works continuously until death during which brain waves are constantly generated according to what we perceive from the environment. By analyzing brain wave patterns, we can predict and identify valuable information on human or animal health. For examples we can monitor coma and brain death in human or animals, various effects of drugs on sleep disorder, day-to-day life human behavior, post-traumatic stress disorders (PTSD), etc. In the experiments conducted, we took the potential differences between the respective channels to identify the variations in brain wave data among the individuals. We used linear interpolation to generate 3D views of the potential data between the locations where the electrodes were placed. A color code is then applied to indicate the range of potential values projected on the human skull. High frequency components were observed near the right parietal and right occipital lobes of the brain. Significant variations were not observed near the frontal or the left region of the brain for a specific activity. The proposed project will introduce a technique to visualize human brain waves in 3D over the skull that will enable us to interpret how these brain waves are associated with various regions on the human brain.
LaSi Spell: Language Agents for Sinhala Spellings
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Samarawickrama, L.; Premarathne, H.L.; De Silva, S.C.M.; Hettige, S.B.
A Spell Checker is a tool which can used as a learning tool for any language learner because meaningful words created through correct spellings. It’s an essential part of several computer softwares such as web browsers, word processors and others because spelling of words give there correct meaning. Sinhala is a language which contain so many spelling rules, while there are few spell checkers has been developed for this language.Multi-agent systems are capable of handling the complexity of the real world problems through its emerging features including communication ,coordination and negotiations. This paper presents design and implementation of multi agent-based spell checker, named LaSi Spell which implemented through the MaSMT framework. It consists of sub ordinary agent systems namely corpus agent,rules agent,gui agent,custom agent and internet agent.LaSi Spell, desktop application has been implemented through java and capable to run with windows. The LaSi Spell has been incrementally tested and has shown encouraging results in its performance.
Use of Image Processing as an Alternative to Manual Traffic Counts
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Herath, O.K.; Sivakumar, T.
Intelligent Transport Systems are essential to achieve efficient and effective traffic management system in the Sri Lankan context. Quality and accurate traffic data are essential in analyzing, visualizing and for future prediction of traffic. We used deep learning based real-time object detection YOLOv3 to traffic surveillance. The research focuses on identifying best camera orientation for better accuracy, transfer-learning of Sri Lankan Vehicles categories into classes, classified vehicle counts using video processing and compare accuracy and efficiency of image processed vehicle classified counts with that of manually collected data. Videos are captured in 1080p @ 60fps at an angle of Ɵ = 0˚ and Ɵ = 15˚ in different heights. Use 500 vehicles in each category to train and 500 vehicles in each category for evaluation. This study intends to apply image processing and Deep Learning based real-time object detector to capture different vehicles classification in order to solve the existing traffic problem in Sri Lanka.
Study on Virtual Learning Environment System in the Field of Construction Technology - A Sri Lankan Universities Perspective
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Charanya, R.; Kesavan, M.
In order to maintain a good relationship in teaching and learning activities among students and university academic staff, a system called Virtual Learning Environment (VLE) can be used. A VLE system was designed among the students and university academic staff members in the field of Construction Technology in Sri Lankan universities to encourage a positive approach in knowledge achievement and to support active learning within the university. This study was carried out to analyze the factors influencing the VLE system and to explore the relationship between the students and university academic staff members on the VLE system. The factors influencing VLE were identified through the literature review and the interviews which were conducted among the university academic staff and the industry experts. A paper-based questionnaire survey was carried out among the students and university academic staff members who used the above created VLE system in the field of Construction Technology in order to measure the severity of the factors influencing the VLE system. There were 40 nos. of responses from the students and 14 nos. of responses from the university academic staff members received. The respondents were requested to indicate their level of contribution on various factors in the survey questionnaire with a 5-point Likert scale. The Relative Importance Index (RII) was calculated for each factor. The severity of each factor was identified based on its RII value. The factors were ranked based on their severity and Spearman’s rank correlation coefficient was calculated. It was found that there was 48.4% of positive degree of agreement between the students and university academic staff on the factors influencing VLE in the field of Construction Technology. The students stated that time saving, infrastructure, collaborative learning, frequent feedback, sustainability and flexible learning are the most significant factors influencing the VLE system, where the university academic staff members identified that collaborative learning, time saving, frequent feedback and infrastructure are the most significant factors influencing the VLE system in the field of Construction Technology from Sri Lankan universities.
Content in Box - Extending Moodle
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Senthuja, K.; Gnanakeethan, B.; Sarveswaran, K.
Aim of this research is to propose a systematic way to share Moodle-based learning materials such as video to remote schools that have limited Internet access. This has become a need in Sri Lanka. Because, serving multimedia contents over the internet is heavily limited by bandwidth and the link speed and this is a common problem everywhere, especially in the rural areas where the penetration of the Internet is very limited. On the other hand, the use of Multimedia has become an inevitable practice in classrooms. Further, self-paced learning has become a focal point in recent time. The government has also taken initiatives to promote self-learning and has invested a lot in setting up the environment and developing content for self-paced learning. In addition, there are studies in the region show that the use of multimedia content would increase the performance of students in national examinations. Shortage of school teachers in remote schools is also another serious issue. On the other hand, now there are several computer laboratories around the nook and corners of the country which are rarely utilized. Some of these laboratories have the Internet, however, students are usually not permitted to use the Internet due to the concern of Internet data cost. Aki.coach has been developed as an online course delivery platform for secondary education in Sri Lanka. To break the obstacles in taking this school level, a portable Aki box is introduced. This box will act as an integrated Content Delivery Network using squid proxy for Moodle. The box comes with all the video and the bandwidth consuming content. If this box is plugged to a laboratory, all the students can do self-paced learning using video and other materials. However, when students access a newly updated video, it will be downloaded via the Internet and stored in the Aki box. Students can also do an online examination. More importantly, students can continue their activities when going home. However, the content will be served from the main server when they access from home which students will feel any differences. Aki box has a Moodle installation which will periodically update the main server. Even if there is an Internet problem still the content can be served from the Aki box and the data will be synced when there is an internet connection. Moodle is altered to always get the video and other bandwidth consuming content from the squid cache if it is available. Now, this box is piloted in one location and soon it will be made available to other schools.
Identification of Papaya Fruit Diseases using Deep Learning Approach
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Munasingha, L.V.; Gunasinghe, H.N.; Dhanapala, W. W. G. D. S.
The diseases are a major problem faced by all the farmers including fruit farmers. It is a threat for large farmlands because these diseases spread throughout the land and make the fruits inedible, which at the end impact badly on the farmer’s income. Hence early disease detection is very important for the farmers to prevent or to control the propagation of the diseases. The traditional method of fruit disease detection and identification is naked eye observation. Even if this method is sufficient for a home gardener, it is a very inefficient one that requires experience and expertise. As a solution for this problem several computerized approaches are being developed using Machine Learning and Image Processing techniques in the resent researches. In our proposed work, we considered Papaya fruit, as it is a very popular fruit cultivation in Sri Lanka. In this study we have implemented a computerized model for papaya disease identification using Convolutional Neural Network (CNN). Among various diseases of papaya fruit, anthracnose, black spot, powdery mildew, phytophthora and ringspot were chosen. These are commonly found in Sri Lankan papaya cultivation. We have collected diseased images using a digital camera in normal conditions from papaya farms. Some of the images were found from the publicly available images on the internet. Then we have trained a deep CNN for these images. The network is able to classify images into five major papaya diseases. The system can finally identify the disease once a new image fed to it. The model performed ~92% of classification accuracy for new images. With compared to previous research done using Support Vector Machine (SVM), there is an increase of ~2%. Furthermore, it could be seen that the Black Spot disease was identified very easily by the model. Powdery Mildew was the most difficult disease to recognize. The results of this study reveal that this method is an accurate, reliable and efficient where it could be useful as an aid for expertise.
Soil Degradability of Food Wrapping Polythene Films Manufactured from PLA, PBAT and LDPE
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Abeywickrama, M.S.J.; Yatawara, M.D.M.D.W.M.M.K.
Food wrapping polythene films manufactured from different virgin materials have become a major environmental concern at present as these films take much time for the complete degradation in the environment. This study assesses the soil degradability of films manufactured from poly-L-lactic acid (PLLA) (28%) + dimethyl ester (38%) + starch (26%) + auxiliaries (8%) (F001), poly-lactic acid (PLA) (F002), poly butylene adipate terephthalate (PBAT) (F003) and low-density polyethylene (LDPE) (F004) in natural soil. The tensile strength, elongation, moisture and water absorption of manufactured films were analyzed at the beginning of the experiment. Half of the manufactured films was immersed in food waste contaminated water and buried at 10 cm depth in soil. The other half was also buried without processing at the same depth in soil. The study was carried out for four months. Percentage degradability was calculated after 02- and 04-month intervals and by using weight losses as a representative parameter of the degradability. Results revealed the significantly highest tensile strength and elongation from manufactured F004. In addition, manufactured F004 showed the significantly lowest water absorption and moisture content (p < 0.05; ANOVA). Nevertheless, the highest percentage degradability (94%) in soil was observed from F003 followed by F002 contaminated with food waste. Moreover, the results showed a poor degradation (< 1%) of films manufactured from F004. The results further revealed that the films contaminated with foods degraded more than films those haven’t contaminated with foods. Therefore, the present study concludes that food wrapping polythene manufactured from PLA and PBAT showed a significant degradation potential within four months whereas films manufactured from LDPE did not show a remarkable degradation within the same time duration.
Integrating Technology into Undergraduate Classroom; Studentled Video Production as an Effective Instructional Strategy
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Rupasinghe, T.P.; Wijesinghe, S.C.
In the current world context, successful integration of digital technology and education theory has led to new advents of teaching and learning. Current students, termed as “Digital natives” have grown up in a multi-media simulated world and are highly competent in adopting to new technologies and therefore, it is importance to utilize technology enabled pedagogical approaches to invoke students’ interest and engagement. Further in the current socio-economic context, it is of enormous importance to enhance students’ generic skills such as self-directed learning, critical thinking, problem-solving, collaboration and cooperation in addition to the domain-specific knowledge and skills. Utilizing digital technologies in the tertiary education can be named as a valuable approach to address above challenges. In the past decade, using digital videos in the teaching and learning have become an emerging instructional strategy, mainly being used in the content delivery. However, there is only a limited number of studies that have been conducted focusing on learning through student-produced digital videos. Current study focuses on investigating student-led video production as an effective active learning instructional strategy. Study was conducted as a part of an Engineering Technology degree program and students (N=72) were asked to create videos (10-15 minutes) to educate their peers on given topics in the course content. Then they were given the opportunity to teach their peers using produced videos. Students’ perception on the activity was evaluated using surveys and its’ impact on the learning process was evaluated through an in-class quiz and was compared with previous quizzes. Majority of the students (> 90 %) had agreed that the activity helped them to understand subject matter better and improved their confidence, communication skills, team work skills and technical skills. Further, according to statistical testing it was proved that the average mark (57 %) for the quiz after the activity was higher than previous quiz (39 %) proving that the activity has a direct impact on students learning. In conclusion, it can be stated that student-led video production has a vast impact as an instructional strategy which enhances students’ competence, generic skills as well as the subject knowledge and thereby enhance the quality of tertiary education.
Automatic Motion Artefacts Recognition in Resting ECG/EKG to Identify Failed Tests using Machine Learning
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Nanayakkara, S.A.; Meegama, R.G.N.
Although an ECG is able to identify certain heart diseases, an uninterrupted and a clear signal is essential to accurately diagnose any abnormalities in the heart functions. Obtaining such a crisp ECG is a challenging task due to several artifacts such as motions because muscle movements are inevitable even in resting ECGs due to medical conditions such as anxiety, Parkinson’s disease and body tremors. In addition, skin stretching too, produces electricity that disturbs the potentials involved in an ECG. There are numerous experiments have been conducted to find effective and efficient motion artifact removal methods from ECGs. In this study, we use cleaned and disturbed ECGs to implement more effective and efficient method to remove motion artifacts and evaluation mechanism for ECGs. The first stage of the proposed technique involved gathering more than 500 ECGs having 12 leads data from public sources available on PhysioNet online database. These data contained cleaned ECGs and disturbed ECGs of healthy and unhealthy patients. The data set is cleaned to remove noise and undesirable effects such as baseline wander. A technique based on multi-resolution thresholding is used to recognize and remove motion artifacts and further, the Savitzky-Golay filter is used to reduce the mean squared error of this process. In the second stage, a convolution neural network (CNN) is implemented on the cleaned ECG dataset. Initially, datasets of 12 leads are shuffled under two categories: with and without noise. These shuffled images, numbering more than 36,000, are then categorized for training, validation and testing of data with and without motion artifacts. Results indicate a 98.7% accuracy in predicting whether a given ECG can be used or not by examining more than 500 cleaned ECGs.
Potential Use of Selected Macrophytes Based Constructed Wetlands for the Treatment of Landfill Leachate
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Perera, K.R.S.; Yatawara, M.D.M.D.W.M.M.K.
Phytoremediation using constructed wetlands (CWs) is widely practiced for the removal of contaminants in landfill leachate. The present study was planned to assess the potential of floating macrophytes (Eichhornia crassipes and Pistia stratiotes) and emergent macrophytes (Typha angustifolia and Chrysopogon zizanioides) in improving the quality of leachate discharged from a Sequencing Batch Reactor (SBR) system located at Dompe sanitary landfill, Gampaha, Sri Lanka. The batch type CWs were arranged to identify the suitable dilution of leachate (as 0%, 25%, 50% and 75%) for the optimum plant growths. Based on the preliminary investigations, the potential of improving leachate quality by (ia) E. crassipes (T501) (ib) T. angustifolia (T502) and C. zizanioides (T503) at 50% dilutions and (ii) T. angustifolia (T04) and C. zizanioides (T05) at 0% dilution were assessed in continuous flow CWs. Water quality parameters including temperature, pH, electrical conductivity, turbidity, BOD, COD, TSS, phosphate, ammonium nitrogen, nitrate, sulphate and color were tested once in five days for 40 days period. Results were subjected to One-way ANOVA followed by Tukey’s pair wise tests in Minitab 14. Two sample t-test at 95% CI was also applied as required. With respect to controls, percentage reductions of measured parameters increased in wetlands having either floating macrophyte, E. crassipes or selected emergent macrophytes. Nevertheless, among the emergent plants, C. zizanioides [(T503) and (T05)] showed the highest performance in improving leachate quality followed by T. angustifolia [(T502) and (T04)]at both dilutions. Although E. crassipes (T501) showed higher percentage reductions of the selected parameters at 50% dilutions, this is not recommended as leachate dilutions are impractical in CWs. P. stratiotes has proven unsuccessful in the present study. Therefore, among the plant species selected, C. zizanioides that performed at 0% dilution could be recommended as the best plant for the remediation of leachate draining from SBR system.
An Unsupervised Machine Learning Approach for Churn Prediction
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Prasanth, S.; Rathnayaka, R.M.K.T.; Arumawadu, H.
Customer churn is one of the critical issues faced by the firms nowadays. Telecommunication industry is no exception to this rule. In this industry, keeping the existing subscriber (customer) is more valuable than acquiring a new subscriber (attracting new customers costs approximately 5 times higher than retaining the existing customers). Therefore, predicting the attrition behavior of customers in advance is a significant task. This behavior has triggered most of the researchers to focus on developing the churn prediction model in several industries. Anyhow, in most of the time supervised machine learning techniques have been incorporated in this regard. But in here, an unsupervised machine learning approach has been proposed. A local telecommunication company can be approached for the purpose of conducting this research. Around 10,000 postpaid subscriber details with 20 attributes have been obtained and analyzed during this research. Further, Principal Component Analysis (PCA) and Kmeans clustering algorithm have been utilized with the intention of reducing the dimensionality between features and to find the churners and non-churners respectively. The results obtained from the PCA have revealed that, 16 principal components which represent all the 20 features are considered as most important aspects to cover the entire data. Moreover, totally 6 clusters have been generated and some particular features that tend to show high contributions were identified during the principal component analysis have been analyzed towards each cluster. The proposed approach has finally revealed that out of the 6 clusters three (3) representing 4888 are churners and the other three (3) representing 5112 are non-churners. It could be ensured that, this approach would assist the future researchers to have a promising start for combining the unsupervised technique with the supervised one.
A Trust based Advanced Machine Learning Approach for Mobile Ad-hoc Network Securty
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Jinarajadasa, G.M.; Liyanage, S.R.
Mobile Ad-hoc Networks (MANETs) are one of the types of Wireless Ad-hoc Networks which consist of autonomous mobile nodes connected wirelessly. These are self-configured, lessinfrastructure networks which are having highly dynamic topologies due to the frequent link changes in the network including the addition of new nodes, removal of existing nodes and etc. Because of this dynamic nature, various issues regarding the reliability of the communication and other security threats such as malicious attacks occur in MANETs. Since ‘Trust’ is the major factor which reliability and the security rely on, enhancing the trust in a MANET ensures that the security of the network environment is achieved. Over the recent past decade, a plenty of researches have been done in the related area including approaches of Machine Learning, Swarm Intelligence, Mobile Agents and Probabilistic Models. When comparing the different properties of each approach such as memory, computational power, flexibility to topology changes, the accuracy of results and cost, applying machine learning techniques has been efficient and accurate in providing results. Among Machine learning approaches reinforcement learning gains a more suitability for applied in mobile ad hoc networks since it gives more accurate results due to the ability to capturing the dynamic behaviour easily as well as no need for historical data to give predictions where it can give predictions on newly joined network nodes also. And when selecting the best algorithm because of the physical distribution of MANET information, an algorithm which has the ability to be distributed among the nodes has to be chosen. Instead of considering direct and indirect trust separately, it is recommended to apply a hybrid trust approach which aggregates the trust values. Hence, considering all this information the future research work is planned to be launch in the area of machine learning; specifically, in the area of reinforcement learning according to the analyzed results of early work. Therefore, this research work is proposing to develop a trust computational model, which uses an advanced machine learning mechanism to predict the trust value of each network node.
Comparing Geant4 Simulated and TALYS-1.8 Code Evaluated Cross-Section Data for 4.438 MeV Gamma ray Line of 12C
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Ramanathan, V.
In present, Monte-Carlo transport code plays a major role in developing detectors, particularly the Geant4 Monte-Carlo code due to its versality and flexibility for different applications. Although, the Geant4 has been an invaluable tool for the development of devices, the discrepancies in prompt gamma cross-section data for prominent elements of human body has been reported in the range of proton therapy (50 -250 MeV). Even though, the binary cascade model has been suggested in proton therapy range, the problems with prompt gamma production cross-section have been reported. The aim of this study is to compare Geant4 simulated and TALYS evaluated prompt gamma cross-section data of 4.438 MeV photo peak of 12C to identify the inconsistency in the cross-section data. TALYS is a nuclear reaction study software which can be used to simulate nuclear reactions in the energy range of 1 keV to 200 MeV. The Geant4 model of AFRODITE detector system has been modeled to mimic the iThemba LABS AFRODITE detector system. The Geant4 AFRODITE model was validated using three standard gamma emitting sources (60Co, 137Cs, and 152Eu). The absolute detector efficiency of the Geant4 AFRODITE model also was determined. In the cross-section measurement simulation study, 1012 proton histories were used to collide the carbon and mylar target in the proton energy range of 66 – 125 MeV. The same procedures were performed experimentally using AFRODITE clover detector system. Further, TALYS 1.8 code was used to simulate the proton interaction with carbon target in the range of 5 to 150 MeV. As with the 4.438 MeV cross-section data comparison, there is a significant inconsistency between Geant4 simulated and TALYS simulation and also with experimental data set. To improve the accuracy of Monte-Carlo simulation study, more experimental cross-section data and the evaluation of proper physics models of Geant4 Monte-Carlo transport code in proton therapy range are future need.
Behavior & Biometrics Based Masquerade Detection Mobile Application
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Chandrasekara, P.; Rajapaksha, S.; Abeywardana, H.; Sanjeevan, P.; Abeywardena, K. Y.
Mobile phone has become an important asset when it comes to personal security since one’s mobile is now a virtual safe for that person. This is due to the sensitivity of the details which are stored in these devices. To protect the information inside a mobile phone the manufacturers use conventional technologies such as password protection, face recognition or finger print protection. Nevertheless, it is clear that these security methods can be bypassed by several other techniques as shoulder surfing, finger print or face recognition by pass with 3D printing. Due to these concerns post authentication is an increasingly discussed topic in the security domain. However, there are very few applied researches done on the post authentication of mobile platforms. In order to protect the phone from an unauthorized user a novel method is proposed by the authors. The aim of the research is to detect the illegitimate user by monitoring the behavior of the user. In order to detect the behavior four main parameters are proposed. Namely, Key stroke dynamics using a customized keyboard, location detection, voice recognition and App usage. Initially machine learning is used to train this mobile application with the authentic user’s behavior and they are stored in a central database. After the initial training period the application is monitoring the usage comparing it with the existing data of the legitimate user. Another unique feature is the inbuilt prevention mechanism which is designed to be executed when an illegitimate user is detected. The entire storage content will be encrypted and a current location alert along with a captured photo of the intruder will be sent to a pre-defined account of the real user in a cloud platform. The real user can log into the account remotely and obtain the phone’s location and the photo of the intruder. Furthermore, this application is proposed as an inbuilt application in order to avoid the deletion of app or uninstallation of the app by the intruder. With this proposed post authentication application “AuthDNA”, a user is able to protect sensitive information of the mobile device in case of theft and bypassing of initial authentication.
High Fidelity Simulation in Undergraduate Medical Curricula: Experience of Fourth Year Medical Students at a Sri Lankan Medical Faculty
(4th International Conference on Advances in Computing and Technology (ICACT ‒ 2019), Faculty of Computing and Technology, University of Kelaniya, Sri Lanka, 2019) Kodikara, K.G.; Karunaratne, W.C.D.; Chandratilake, M.N.
Application of theoretical knowledge to management of critically ill patients is a challenging task faced by medical undergraduates where opportunities to learn clinical skills with regard to management of emergencies are few. High fidelity simulation (HFS) is widely used globally as an adjunct to clinical practice enabling students to learn clinical skills in a safe environment. However, research in the use of HFS in Sri Lanka is minimal. The purpose of this study was to explore the response of medical undergraduates to a high-fidelity simulator (HFS) in the context of management of emergencies. A pilot group of 30 fourth year medical students underwent a high-fidelity simulator session. They completed a self-administered evaluation, which included both open and close ended questions and participated in a focus group discussion post-simulation. Descriptive statistics were employed to analyze the responses to close-ended questions and the responses of the focus group discussion and open-ended questions were analyzed for recurring themes. All participating students responded to the evaluation. Students rated the simulation-based learning experience with high positivity. The self-competency of 29 (96.6%) students had increased following the sessions. The session provided a safe learning environment to all students. 19 (63.3%) students felt it helped put theory into practice while 21 (70.7%) students identified it as good practice for internship. 25 (83.3%) students wished to participate in more sessions. 17 (56.6%) students commented on the realistic nature of the experience. This study confirmed findings of previous studies conducted using HFS among medical undergraduates, confirming that the students highly valued high-fidelity simulation and find the opportunity to apply theoretical knowledge to practice in a safe environment. A high-fidelity simulator is a valuable learning tool in undergraduate medical education.