Textual information is becoming increasingly available with the expansion of the World Wide Web and the availability of the mass storage devices. As a result, more advanced techniques are needed to retrieve only the information that is relevant to the user.;The principal way of retrieving information is to use a query which consists of keywords/phrases that reflect the information needs of a user. The goal of this thesis is to explore novel means of identifying keywords/ phrases. Here we propose a query expansion technique which not only depends on the underlying content of the documents but also depends on the time period of the documents as well. We introduce two methods, temporal associations and temporal synonyms to discover additional query expansion terms according to the original query of the user. These methods will facilitate the query expansion process and also assist in studying the progression of a story.;Application of our algorithms to the document sets such as Reuters news articles and PubMed abstracts show that an increase in retrieval performance can be obtained using the proposed techniques.
展开▼
