Predicting Search Intent Based on Pre-Search Context

Kong, Weize; Li, Rui; Luo, Jie; Zhang, Aston; Chang, Yi; Allan, James

doi:10.1145/2766462.2767757

Cited by 35 publications

(42 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to Cheng et al, many searches are triggered by browsed web pages [42]. Kong et al tried to predict search intent using recently browsed news article before search [43]. A large number of queries are triggered by news article daily [43].…”

Section: User Intentionmentioning

confidence: 99%

“…Kong et al tried to predict search intent using recently browsed news article before search [43]. A large number of queries are triggered by news article daily [43]. Predicting search intent using browsed pages is inadequate [43].Our proposed method uses live RSS newsfeed for query prediction.…”

Section: User Intentionmentioning

confidence: 99%

“…A large number of queries are triggered by news article daily [43]. Predicting search intent using browsed pages is inadequate [43].Our proposed method uses live RSS newsfeed for query prediction. It makes use of user profiles to predict the search intent.…”

Section: User Intentionmentioning

confidence: 99%

See 2 more Smart Citations

Context Sensitive Search String Composition Algorithm using User Intention to Handle Ambiguous Keywords

Gajendragadkar

Joshi²

2017

IJECE

View full text Add to dashboard Cite

Finding the required URL among the first few result pages of a search engine is still a challenging task. This may require number of reformulations of the search string thus adversely affecting user's search time. Query ambiguity and polysemy are major reasons for not obtaining relevant results in the top few result pages. Efficient query composition and data organization are necessary for getting effective results. Context of the information need and the user intent may improve the autocomplete feature of existing search engines. This research proposes a Funnel Mesh-5 algorithm (FM5) to construct a search string taking into account context of information need and user intention with three main steps 1) Predict user intention with user profiles and the past searches via weighted mesh structure 2) Resolve ambiguity and polysemy of search strings with context and user intention 3) Generate a personalized disambiguated search string by query expansion encompassing user intention and predicted query. Experimental results for the proposed approach and a comparison with direct use of search engine are presented. A comparison of FM5 algorithm with K Nearest Neighbor algorithm for user intention identification is also presented. The proposed system provides better precision for search results for ambiguous search strings with improved identification of the user intention. Results are presented for English language dataset as well as Marathi (an Indian language) dataset of ambiguous search strings. Keyword: Autocompletion Context Data mining Search User intention Copyright © 2017 Institute of Advanced Engineering and Science.All rights reserved. Corresponding Author:Uma Gajendragadkar, COEP, Phone +919822479128, G7/9 Omkar Garden, Manikbaug, Pune, Maharshtra, India. Email: umagadkar@gmail.com INTRODUCTIONCurrent search engines churn a large volume of data to obtain meaningful information; however, the main challenge is to get relevant results in the top few result pages [1], [2]. Search engines check for the presence of keywords in documents. Mere presence of keywords in a document may not match the user's search intention and need. User satisfaction increases when more relevant and exact information is presented in the top few results. An appropriately composed query is the starting point for handling this challenge [3]. Performance of search engines can be improved with the use of appropriate keywords or prediction of such keywords [4][5][6]. Search engines use search logs and most popular queries; however, these are not sufficient to predict the user's interests or intention [7].Users are of three types, first -Internet skilled users, second -Internet aware users and thirdInternet unskilled users. Many times, users do not know the proper keywords for searching information and they cannot express their information need or intent of search [8], [9]. This results in search results often not satisfying user's information need. This problem can be addressed by query expansion and reformulation [3]. Search engines provide a...

show abstract

Section: User Intentionmentioning

confidence: 99%

Section: User Intentionmentioning

confidence: 99%

See 1 more Smart Citation

Context Sensitive Search String Composition Algorithm using User Intention to Handle Ambiguous Keywords

Gajendragadkar

Joshi²

2017

IJECE

View full text Add to dashboard Cite

show abstract

“…The relevance scores evaluated in this work make use of the existing research, such as MPC [3,19,24,34], Personal(-S) [3,6,34], and TimeSense(-S) [6,35,37,29]. More recent QAC methods also predicted the likelihood that suggested queries would be selected by users based on keystroke behaviors during query compositions [24,43,23], determined suggestion rankings based on query reformulation signals [19], exploited web content signals [22], or combined signals such as time and previous queries from users [6]. Specifically, Zhang et al proposed adaQAC, an adaptive QAC model incorporating users' implicit negative feedback [43].…”

Section: Related Workmentioning

confidence: 99%

Towards Mobile Query Auto-Completion

Zhang

Goyal

Baeza-Yates

et al. 2016

Proceedings of the 25th International Conference on World Wide Web

Self Cite

View full text Add to dashboard Cite

We study the new mobile query auto-completion (QAC) problem to exploit mobile devices' exclusive signals, such as those related to mobile applications (apps). We propose AppAware, a novel QAC model using installed app and recently opened app signals to suggest queries for matching input prefixes on mobile devices. To overcome the challenge of noisy and voluminous signals, Ap-pAware optimizes composite objectives with a lighter processing cost at a linear rate of convergence. We conduct experiments on a large commercial data set of mobile queries and apps. Installed app and recently opened app signals consistently and significantly boost the accuracy of various baseline QAC models on mobile devices.

show abstract

“…Kong et al (2015) proposed two types of short-term context: pre-search context and in-search context, and suggested that pre-search context triggers the search. Liu and Wu (2015) suggested that contextual information, which included time, weather, and emotions, influences a user's preference on a certain point of interest.…”

Section: Research On the Mobile Search Contextmentioning

confidence: 99%

How Users Search the Mobile Web: A Model for Understanding the Impact of Motivation and Context on Search Behaviors

Zhu

Ran

2016

Journal of Data and Information Science

View full text Add to dashboard Cite

Purpose: This study explores how search motivation and context influence mobile Web search behaviors.Design/methodology/approach: We studied 30 experienced mobile Web users via questionnaires, semi-structured interviews, and an online diary tool that participants used to record their daily search activities. SQLite Developer was used to extract data from the users' phone logs for correlation analysis in Statistical Product and Service Solutions (SPSS). Findings:One quarter of mobile search sessions were driven by two or more search motivations. It was especially difficult to distinguish curiosity from time killing in particular user reporting. Multi-dimensional contexts and motivations influenced mobile search behaviors, and among the context dimensions, gender, place, activities they engaged in while searching, task importance, portal, and interpersonal relations (whether accompanied or alone when searching) correlated with each other. Research limitations:The sample was comprised entirely of college students, so our findings may not generalize to other populations. More participants and longer experimental duration will improve the accuracy and objectivity of the research. Practical implications:Motivation analysis and search context recognition can help mobile service providers design applications and services for particular mobile contexts and usages.Originality/value: Most current research focuses on specific contexts, such as studies on place, or other contextual influences on mobile search, and lacks a systematic analysis of mobile search context. Based on analysis of the impact of mobile search motivations and search context on search behaviors, we built a multi-dimensional model of mobile search behaviors.

show abstract

Predicting Search Intent Based on Pre-Search Context

Cited by 35 publications

References 26 publications

Context Sensitive Search String Composition Algorithm using User Intention to Handle Ambiguous Keywords

Context Sensitive Search String Composition Algorithm using User Intention to Handle Ambiguous Keywords

Towards Mobile Query Auto-Completion

How Users Search the Mobile Web: A Model for Understanding the Impact of Motivation and Context on Search Behaviors

Contact Info

Product

Resources

About