Analysis of Interaction Patterns and Users' Query Reformulation Strategies in Persian Search Engine

Web search is integrated into the modern online life, which is partly represented in users’ queries[p1] [F.2] . Search engines receive users' queries and then retrieve a limited number of relevant documents from a pool of billion web pages. Therefore, they can reveal users’ behavior patterns using their search logs in long periods of time. Behavior patterns can be identified in query expansion, query suggestion and spelling correction. In the present study, query logs of Parsijoo Persian search engine were examined to reveal users’ interaction pattern(s), and specifically, lexical and temporal properties of the query logs, as well as reformulation patterns. Query term distribution, temporal query length variation in different hours of the day and patterns of users’ query reformulation(s) were analyzed. The results of analyzes were employed to investigate users' web search behaviors in contrasts to behaviors reported for human users in studies which used search logs of the international search engines. The results of the present study demonstrated that the Persian native users show similar behavior patterns