The Web Science Trust

Discovering Social Relationships and Intentions in Web Forums Using Natural Language Processing Techniques

Casillas, Luis and Daradoumis, Thanasis (2009) Discovering Social Relationships and Intentions in Web Forums Using Natural Language Processing Techniques. In: Proceedings of the WebSci'09: Society On-Line, 18-20 March 2009, Athens, Greece. (In Press)

[img]
Preview
PDF (Poster Description) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
59Kb

Abstract

Natural Language (NL) is the main communication media for humans, who have a deep need for communicating based on evolutionary mechanisms. This inner-need implied the development of languages. Proto-humans had to name objects around, feelings, memories, expectations, and etcetera. The growing of conscience in human’s brain along the time, led to the construction of complex models of concepts stored among the neurons of the brain. Some layers of the brain were devoted to handle the relations between an inner net of concepts with the areas concerning the treatment of language. The NL spoken by humans has a complex construction, based on the diversity of meaning for words and expressions according to the context. Besides, there are different cases of languages. Our work is aimed at presenting a multi-agent approach for dealing with the problem of discovering the intention in expressions written in Spanish and inside Web forums. We are also looking for the identification of some relations among participants in the Web forum. Although the conversations in these forums are usually built in some point between formal and informal communications, and would not follow grammatical rules in every case, we have made arrangements to deal with many of the situations present in Web forums and provide an adaptive response for most of the cases.

Item Type:Conference or Workshop Item (Poster)
Uncontrolled Keywords:Conversation Analysis, Web Forums, Natural Language Processing, Intention Detection
Subjects:Web Science Events > Web Science 2009
ID Code:111
Deposited By: W S T Administrator
Deposited On:24 Jan 2009 08:45
Last Modified:25 Oct 2011 16:33

Repository Staff Only: item control page

EPrints Logo
Web Science Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.