This is a great topic for a graduate thesis: Traditional IR considers the problem of matching documents to a query as a single information need to be satisfied. However, since a system doesn’t know what is in the mind of users, the query itself can be a multiple information need.
When you think thoroughly this is why Web searching, how users search and reformulate queries on the Web, is different from, for example, IR searching –wherein user resource to query expansion and relevance feedback mechanisms.
In Query Phrase Suggestion from Topically Tagged Session Logs, the authors addressed this. They wrote:
“The definition of relevance in a phrase suggestion task is different from that in the information retrieval problem. Rather than assuming that the query represents a single information need and seeking to find documents relevant to that need, the suggestion task assumes a query could represent multiple information needs.”
In the January 2007 issue of IR Watch – The Newsletter we discussed this topic in details.
Still not a subscriber of IRW? Why not?
This post is based on a legacy post originally published in 12/30/2006.