Analysis of Instant Search Query Logs

Instant search is a new search paradigm that shows results as a user types in a query. It has become increasingly popular in recent years due to its simplicity and power. In an instantsearch system, every keystroke from a user triggers a new request to the server. Therefore, its log has a richer content than that of a traditional search system, and previous log analysis research is not applicable to this type of log. In this paper, we study the problem of analyzing the query log of an instant-search system. We propose a classification scheme for user typing behaviors. We also compare the log of an instant-search system and that of a traditional search system on the same data. The results show that on a people directory search system, instant search can typically save 2 seconds per search, reduce the typing effort by showing the results with fewer characters entered, and increase the success rate.