Date Added: Apr 2012
The paper aims at improving domain specific search engine which are becoming more popular as compared to web-wide search engines as they are difficult and time consuming to maintain. At the same time they are unable to provide sufficient relevant documents to represent the target text. The paper addresses the problem of generating topic hierarchies for diverse text segments with a general and practical approach that uses the web as an additional knowledge source. Unlike long documents, short text segments typically do not contain enough information to extract reliable features. The paper investigates the possibilities of using highly ranked search result snippets to enrich the representation of text segments.