Browsing by Author "Sarveswaran, K."

Now showing 1 - 3 of 3

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.
Recognising Elders using Behavioural Biometrics
(Faculty of Computing and Technology, University of Kelaniya, Sri Lanka., 2017) Preethiraj, R.; Sarveswaran, K.
The elderly population continues to grow everywhere and it finds difficulties to access websites due to various reasons including functional impairments like lack in vision, hearing, mobility and movement. Therefore, websites are usually made separately for elders to improve their user experience. However, first it’s important to recognise whether a user is an elder or not, and for that usually user profile information such as date of birth or age are used. Users may reluctant to feed information or may even feed a wrong one. This research proposes a method using which elders can automatically be recognised using behavioural biometrics of them. Based on the initial observational study on elders it was noted that elders shake the mouse to identify the mouse pointer location, do scrolling fast without much control, and the elders take a lot of time to click on a link or button after moving over it. These three observation were considered as behavioural biometrics to recognise elders. A data set was compiled in a control environment from 24 people of different ages including 18 elders who are more than age of 65. All the people were asked to follow a same set of tasks in two websites. Thereafter, the collected data were cleaned and a decision tree was built to recognise elders using j48 algorithm and Weka tool. The results showed that elders move the mouse faster than 5.7928 pixel/millisecond, scroll faster than 3.455561/millisecond, and take more than 1, 158.6875 milliseconds to respond over a link or button. Thereafter more behavioural biometrics were collected from random users in open environment in which users were asked to fill a questionnaire with the intention of collecting their age. The collected data then were used to validate the decision tree. It was found that speed of mouse movement recognises the elders with 84.51% accuracy, scrolling speed recognises with 96.08% accuracy, and response time recognises elders with accuracy of 97.68%. The results show that instead of rely on user profiles, elders can be recognised using user behavioural biometrics with significant accuracy. Though the response time shows a high recognition rate, it is planned to explore the combination of different behaviour biometrics together to see whether recognition rate can be improved.
SATS: A Computer Aided Translation Engine.
(The Third International Conference on Linguistics in Sri Lanka, ICLSL 2017. Department of Linguistics, University of Kelaniya, Sri Lanka., 2017) Kumara, I.; Mohomad, N.; Sarveswaran, K.
Computer used language translation is a controversial subject. Even though many research studies have been done in this area of study, still there are no machine translation systems available for Sinhala – Tamil – English translations with a reportable accuracy . The complexity of Sinhala and Tamil languages makes the language processing, specifically machine translation a difficult task. Among the available tools for translations, Google translation engine performs better and it also has a wider usage. However, it has weaknesses like regional slangs specifically for Tamil. The system is tuned more towards Indian Tamil slang, because, the Tamil content that were used to train the Google Translator were created mostly by Indian Tamil speakers. This research proposes a mechanism which can be used professional language translations with the aid of computers. The proposed method has a Statistical Machine Translation (SMT) engine, to assist human translators for the translation, which is built using Moses as a framework with giza++ for word alignment and IRSTLM for language modelling. When a user requests for a translation, which can be a word, phrase or small paragraph, it will be sent to SMT engine first. Engine will generate suggestions and will send to a professional translator among the SATS (Sinhala and Tamil Speakers) group automatically along with the request. Next the translator makes correction or approves the translation suggested by the SMT. Thereafter it will be sent to the requested user. More importantly, the translation which was approved by translators will also be stored in a Translation Memory (TM) so that in future, translations can be pulled from TM if someone makes the same request for translation. The system provides accurate translations due to the involvement of experts. However, with the time the system will evolve as an expert language translation system.