Relevant term suggestion in interactive web search based on contextual information in query session logs

Abstract

This paper proposes an effective term suggestion approach to interactive Web search. Conventional approaches to making term suggestions involve extracting co-occurring keyterms from highly ranked retrieved documents. Such approaches must deal with term extraction difficulties and interference from irrelevant documents, and, more importantly, have difficulty extracting terms that are conceptually related but do not frequently co-occur in documents. In this paper, we present a new, effective log-based approach to relevant term extraction and term suggestion. Using this approach, the relevant terms suggested for a user query are those that cooccur in similar query sessions from search engine logs, rather than in the retrieved documents. In addition, the suggested terms in each interactive search step can be organized according to its relevance to the entire query session, rather than to the most recent single query as in conventional approaches. The proposed approach was tested using a proxy server log containing about two million query transactions submitted to search engines in Taiwan. The obtained experimental results show that the proposed approach can provide organized and highly relevant terms, and can exploit the contextual information in a user's query session to make more effective suggestions.

Query Generation as Result Aggregation for Knowledge Representation

“…Smarter approaches have incorporated more session context than this, for substantial gains. Such approaches, for instance, learn a Markov transition model of queries from logs, calculating the probability that pairs of queries co-occur within a portion of a search session [8,13,14]. Others generalize this approach to sequences of n queries [4,12] and have been shown to outperform pairwise methods.…”

A query suggestion method combining TF-IDF and Jaccard Coefficient for interactive web search

“…[1,4,8,[11][12][13][14][15][16][17][18] Various ontologies have been applied to create knowledge-driven models for generating query suggestions, such as WordNet, [19,20] Wikipedia, [21] ODP and YAGO. [22][23][24][25] Query suggestions can also be generated from query related features extracted from web documents returned by search engines.…”

Inferring User Goals Using Customer Feedback andAnalyzing Customer Behavior

“…Here the feedback sessions are formed with the series of both clicked and unclicked URLs and ends with the last URL that was clicked in a session from user click-through logs. [4], [8], [9], [12], [15] demonstrated the use of logs. In search engine the user goal can also be inferred by using the clickthrough data [7], [10].…”