K., Raghuveer and G., Raghavendra Rao and G. L., Shekar and R, Dayanand (2006) An approach for selecting text from e-mail in satellite-based education. In: Knowledge Management International Conference and Exhibition 2006 (KMICE 2006), 6-8 June 2006, The Legend Hotel Kuala Lumpur.

PDF Restricted to Registered users only Download (45kB)

Abstract

It was in September 2004 that Indian Space Research Organization(ISRO), in its endeavor to provide sustainable distance education service in India, launched EDUSAT that is designed to supplement the present teaching system in the country.The EDUSAT project has sought to wipe out disparities in quality between the various educational institut ions by addressing itself to key issues of access, interactivity, cost-effectiveness and consistency of information to students.One subject expert can simultaneously teach hundreds of students in multiple locations across a vast geographical area.The major issue of paucity of faculty members in higher education has been tackled in one single stroke. In this regard, VTU has set up a Forum(http://forum.vtu.ac.in), which provides a platform for students to exchange their view, share resources and also get answers from subject experts by sending their queries into this Forum by e-mail.As this VTU-EDUSAT network is being used by a whopping 150,000 students, there will be thousands of questions from students across different subjects that will be handled through the network by hundreds of subject experts.The questions will be aired by the students during the live sessions of the satellite-based programme or by sending e-mail post transmission of sessions.Further, there will be number of questions by the students after they go through the e-Learning content.It has become impossible to answer all these questions by the subject experts because of paucity of time.It is also observed that there will be many questions that are similar or repetitive or these questions would have been answered earlier by the subject experts.Also, it is impossible for the subject expert to go through all the e-mails completely.Therefore, it is required to develop a solution that involves summarization of e-mail text sent by the students, which means selection of important sentences or text so as to compare them with the current database of questions. Once comparison is made, only the non- repetitive questions are extracted and place them before the subject expert. Further, if the answer is already available for questions, the system advices students to refer the FAQ data base.Here, it is necessary to distinguish the important information with less important information within the original text (e-mail).For this, we need the combination of two steps: information filtering and text reduction.The on-going research is to address such an issue by domain-specific text summarization. We have made an attempt to develop a model that examines content-based text selection.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	ISBN 983-3282-90-3 Organized by: Faculty of Information Technology, UUM
Uncontrolled Keywords:	Information retrieval, Text mining, Domain specific summarization
Subjects:	T Technology > T Technology (General)
Divisions:	College of Arts and Sciences
Depositing User:	Mrs. Norazmilah Yaakub
Date Deposited:	14 Jul 2014 07:01
Last Modified:	14 Jul 2014 07:01
URI:	https://repo.uum.edu.my/id/eprint/11520

Actions (login required)

View Item