https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4142 2026-03-29T10:08:50Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4818 Title: Impact of video surveillance system on ATM PIN security Authors: Perera, D.; Samarasekara, H.; Seneviratne, P. Abstract: Abstract In Sri Lanka, it is a common practice to install wall mounted surveillance cameras in ATM cubicles. Surveillance cameras are installed to detect and/or monitor the situation and people inside the ATM cubicle for safety reasons and as a precautionary measure against the disputes regarding the dispense of cash. However, the researcher show that ad-hoc implementation of surveillance cameras inside an ATM cubicle has a potential threat to ATM PIN security. In the course of the background study the researchers identi ed that, in some cases, the PIN entering process is clearly visible through the surveillance camera footage. Researchers also show that it is possible to infer the ATM PIN simply by observing the forearm movement pattern even when the PIN pad is not visible on the video footage. In most banks, the surveillance camera footage are available to the personnel who are not cleared to access the banking system. In addition, banks operate under the premise that the PIN number is known only to the customer. Elaborate systems, such as HSMs and PIN Mailers, are in place to ensure that even the banking sta is not privy to the PIN number of a customer. Therefore, the potential exposure of the PIN number of a customer to a third party through the CCTV camera footage can be considered as a severe security violation in the electronic banking system. For this PIN inference process researchers have developed a programme using OpenCV and Python. Furthermore researchers have have propose a novel method to identify the key press events using the gradient of forearm movements. 2020-02-01T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4817 Title: Summarization of Stock Market Investment News Articles For Stock Traders Authors: Logeesan, J.; Rishoban, Y. Abstract: Abstract Stock market word itself is something powerful in the Finance world. Field which is used to generate more profit for the organization and individuals. Movement of the stock market can be predicted using different factors such as past stock numeric data, stocktwits, company's portfolio, financial news articles etc. There are researches that were carried using stock numeric data to predict the movement of stock, but only numeric data itself is not possible to predict the movement without the real time information about the organization. If we want to gain real time information about the organization, we must gather information from the financial news articles. Stock traders aren't able to read many news articles within a small time to gain more information unless it is in the summarized form. If traders can read a lot they will get a lot of information and profit gaining may increase. To make the traders read many articles, we need to provide the summarization tool that gives the summary of the significant contents needed for the stock traders. So we carried out this research to identify how far the summarization of stock market news articles will be helpful for the stock traders to carry out their trading activities. We carried this research by collecting news articles from the popular websites and collecting keywords which are used by the stock traders to read the news articles, then pre processing was done on the news articles , then sentences were weighted based on the keywords and the graph analysis is performed to retrieve the salient sentences for the stock traders from the news articles. Generated summary was compared with the summary of the domain experts summary for the corresponding article. Accuracy of the model was more than 55% for every article. So can conclude that summarization of stock market news articles is helpful for the stock traders to carry out their trading activities We believe that, this is the first time, the summarization techniques have been applied specifically in the stock market investment news articles in such a way it is helpful for the stock traders to carry out their trading activities. 2020-01-01T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4816 Title: Landslide Susceptibility Prediction Model using Random Forest for Kalutara District, Sri Lanka Authors: Liyanage, L.C.; Palliyaguru, S.T.; Weerakoon, O.S. Abstract: Abstract Landslides are one of the most recurrent and prominent natural disasters in Sri Lanka. An area of nearly 20,000 sq. km encompassing 10 districts is prone to landslides. According to statistics provided by the National Building Research Organization landslides have destroyed over 800 lives in Sri Lanka over the last decade. In 2017 Kalutara district reported the maximum number of deaths of 101 due to landslides. Owing to haphazard, unplanned land use, inappropriate construction methods, wanton human intervention and other other geological and morphological causes, the trend of landslide occurrence will continue in the next eras. Therefore prediction of landslide susceptibility is indispensable for disaster management and ensure sustainability of developments. The main focus of this study was to investigate the applicability of 12 landslide conditioning factors including slope, aspect, hydrology, Stream Power Index(SPI), Topographic Wetness Index(TWI), Sediment Transport Index(STI), geology, land form, land use, soil type, soil thickness and rainfall in the prediction of landslide susceptibility in Kalutara district using Random Forest machine learning algorithm. In order to achieve this a Geographical Information System(GIS) was used to manipulate and analyze the spatial data while the implementation of the prediction model was carried out using python. A pilot study was carried out to analyze the correlation between the landslide conditioning factors and landslide occurrence and to select the most appropriate set of conditioning factors for the prediction. A landslide inventory of 84 landslides occurrences in Kalutara district, was utilized along with randomly generated 84 non-landslide locations from the landslide-free area of Kalutara district. Random Forest (RF), a non-parametric supervised classification algorithm was employed to construct the prediction model. The efficiency of the Random Forest model was evaluated using Receiver Operating Characteristic(ROC), accuracy, sensitivity and specificity. The results indicated 76.92% specificity, 84.00% specificity, and accuracy of 80.39%. The area under the ROC curve demonstrated 79.46% of predictive capability for the model. Keywords: Landslide Susceptibility, Machine Learning, GIS, Random Forest 2020-02-01T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4815 Title: The Use of Conversational Interfaces in Long Term Patient Care Authors: Gunathilaka, L.A.S.M.; Weerasinghe, W.A.U.S.; Wickramasinghe, I.N. Abstract: Abstract Communication plays a significant role in human life. To fulfill the different requirements people use many communication methods. After learning their mother tongue they use native or foreign languages to communicate. People use different communication methods such as verbal, written, visual, paralanguage and other methods. However, people with disabilities face many problems when communicating with others. Even the patients with long term illnesses who could not complete their work alone expect others to listen to them. By the way, it is a necessity to have a caretaker to take care of the long term patients. With the increasing number of the elderly and disable people, it is essential to use the technology to fulfill their needs and make them more independent. In most of the countries, they spend a more independent life thanks to technology and a comfortable lifestyle. In Sri Lanka, the elderly people get the assistance of a caretaker, since they are not very familiar with the technology and due to language barriers. The machines emulate conversations with humans through conversational interfaces. Mainly we can identify chatbot as a main conversational interface where it’s available on-screen and as a voice assistant. Since the voice integrated personal assistance is more fitting to the patient care domain, the thesis presents the observations done with the long term patients and the elders on their ability to communicate voice integrated personal assistance. The research is conducted using the design science research methodology to understand the current problem in society. The research is initiated with the motivation to facilitate the communication and psychological requirements of long term patients. For that, researchers conducted a review analysis on voice integrated personal assistance and the use of it in the long term patients and elderly people. Then the interview is conducted to identify the requirements of the elderly long term patients. By the results of the review analysis and the interview process, we prioritized the communication needs of them. Using the voice integrated personal assistance the proof of concept was developed for the identified needs. The proposed proof of concept is used to evaluate it on the elderly long term patients by mainly considering requirements such as loneliness, exercise tracking, and games. The Developed solution quantitatively evaluated by elderly long term patients mainly under five criteria such as google home device, mind game, bedridden exercises, talking companion and overall conclusion. Most of them had a positive opinion to embrace the new experience while few of them had language issues due to the accent. In the analysis of overall satisfaction, more than 90% had total satisfaction of 0.75 or more which shows higher satisfaction. Elderly patients believe the device improves their independence and helps them to spend some quality time. More than using the screens to communicate, voice integrated personal assistance is a new experience that makes their life easy. Based on the results, researches presented appropriateness of the voice integrated personal assistance to the elderly long term patient care domain. 2020-02-20T00:00:00Z