Description: Wikipedia is a freely available online encyclopedia developed by a community of users. This encyclopedia comprises of millions of articles. The depth and coverage of Wikipedia has attracted the attention of researchers for employing it as a knowledge resource for solving various problems. In this research we propose to exploit Wikipedia along with other related open knowledge sources to automatically generate Semantic knowledge. We discuss Wikipedia’s structure in detail and suggest hybrid approaches utilizing ontological, structured, semi-structured and unstructured information derived from Wikipedia and similar knowledge sources. We plan to demonstrate the value of the derived semantic knowledge by developing problem specific knowledge based approaches targeting at a set of diverse use cases: namely, document concept prediction, information retrieval, entity classification and Entity Co-reference resolution. Wikipedia has millions of articles and is growing continuously, using it in real world scenarios poses many challenges related to keeping the derived knowledge up to date with Wikipedia. We also propose to engineer an efficient, scalable and evolving architecture that would evolve along with the available online Wikipedia. (PhD dissertation proposal)

Type: Presentation

Date: February 06, 2009

Tags: wikipedia, semantic web, information retrieval, information extraction

Format: Microsoft PowerPoint (Need a reader? Get one here)

Number of downloads: 296

Access Control: Publicly Available

Available for download as

size: 1305088 bytes

Assertions:

1. (Resource) Wikitology: A Wikipedia Derived Knowledge Base is the PowerPoint slides of (Event) Wikitology: A Wikipedia Derived Knowledge Base