Beyond Click Graph: Topic Modeling for Search Engine Query Log Analysis

Knowledge-Based Systems

Yang

et al. 2015

Self Cite

Section: Quantitative Evaluation Of Upmmentioning

confidence: 99%

Query suggestion with diversification and personalization

Knowledge-Based Systems

Yang

et al. 2015

Self Cite

“…The first baseline is the widely used Latent Dirichlet Allocation (LDA) [19], the second one and third one, PTM1 and PTM2 are designed for search engine query log analysis [21], the fourth one is the Topic-Over-Time (TOT) model [29]. We also compare UPM with the Meta-word Model (MWM), the Term-URL Model (TUM), the Clickthrough Model (CTM) [34] and the SSTM model [35]. We compare the strength of the baseline models with the UPM in terms of how well the models can predict the remaining query words after observing a portion of the user's web search history.…”

Section: ) Quantitative Evaluation Of Upmmentioning

confidence: 99%

Personalized Query Suggestion With Diversity Awareness

2014 IEEE 30th International Conference on Data Engineering

Vosecky

et al. 2014

Self Cite

Query suggestion is an important functionality provided by the search engine to facilitate information seeking of the users. Existing query suggestion methods usually focus on recommending queries that are the most relevant to the input query. However, such relevance-oriented strategy cannot effectively handle query uncertainty, a common scenario that the input query can be interpreted as multiple different meanings. To alleviate this problem, the concepts of diversification and personalization have been individually introduced to query suggestion systems. These two concepts are often seen as incompatible alternatives, because diversification considers multiple aspects of the input query to maximize the probability that some query aspect is relevant to the user while personalization aims to adapt the suggestions to a specific aspect that aligns with the preference of a specific user. In this paper, we refute this antagonistic view and propose a new query suggestion paradigm, Personalized Query Suggestion With Diversity Awareness (PQS-DA) to effectively combine diversification and personalization into one unified framework. In PQS-DA, the suggested queries are effectively diversified to cover different potential facets of the input query while the ranking of suggested queries are personalized to ensure that the top ones are those that align with a user's personal preference. We evaluate PQS-DA on a real-life search engine query log against several state-of-the-art methods with respect to a variety of metrics. The experimental results verify our hypothesis that diversification and personalization can be effectively integrated and they are able to enhance each other within the PQS-DA framework, which significantly outperforms several strong baselines with respect to a series of metrics. I. INTRODUCTIONQuery suggestion is an important functionality provided by contemporary web search engines to help users formulate more effective search queries [1]. Most of the existing query suggestion methods [2][3][4][1] [5] belong to the category of relevance-oriented query suggestion, which focuses on maximizing the overall relevance of the suggested queries in response to an input query. However, since search queries are typically short and ambiguous [6], the simplistic relevanceoriented methods usually fail in the face of query uncertainty, which widely exists in the scenario of general web search.Consider the following example of query uncertainty: When the search query "sun" is submitted to the search engine, the underlying information need can be related to at least one of the three facets: the star of the solar system, the computer manufacturer named Sun Microsystems or a newspaper in the United Kingdom. In this case, the relevance-oriented approaches usually generate suggestions that cover a few

show abstract

“…Blei et al [1] proposed Latent Dirichlet Allocation (LDA) to analyze electronic archives. Topic models were since applied in various domains, including search query logs [6,7] and app marketplaces [8]. In the microblogging environment, Hong et al [5] study approaches to apply LDA on microblog data.…”

Section: Related Workmentioning

confidence: 99%

Dynamic multi-faceted topic discovery in twitter

Vosecky

Proceedings of the 22nd ACM International Conference on Conference on Information &Amp; Knowledge Management - CIKM '13

et al. 2013

Self Cite