Predicting Search Intent Based on In-Search Context for Exploratory Search

Predicting Search Intent Based on In-Search Context for Exploratory Search

Vikram Singh (National Institute of Technology, Kurukshetra, India)
DOI: 10.4018/IJAPUC.2019070104

Abstract

Modern information systems are expected to assist users with diverse goals, via exploiting the topical dimension (‘what' the user is searching for) of information needs. However, the intent dimension (‘why' the user is searching) has preferred relatively lesser for the same intention. Traditionally, the intent is an ‘immediate reason, purpose, or goal' that motivates the user search, and captured in search contexts (Pre-search, In-search, Pro-Search), an ideal information system would be able to use. This article proposes a novel intent estimation strategy; based on the intuition that captured intent, and proactively extracts likely results. The captured ‘Pre-search' context adapts query term proximities within matched results beside document-term statistics and pseudo-relevance feedback with user-relevance feedback for In-search. The assessment asserts the superior performance of the proposed strategy over the equivalent on tradeoffs, e.g., novelty, diversity (coverage, topicality), retrieval (precision, recall, F-measure) and exploitation vs exploration.
Article Preview
Top

Introduction

Search systems are used to discover information that aids us in our routine tasks, be they leisure or professional. An ideal information system would support the user in exploiting useful information that can be used in steering these primary tasks the user is initiating (Vakkari, 2001). A primary-task can be any cognitively multifaceted practice performed by humans that may invoke information needs- the real-world task, which is the main task one has to carry out (Belkin & Croft, 1992). For example, a user may search ‘Blockchain’ because of a buzzing around, but the system does not know which of the pre-search contexts (the news article) is the prompt cause.

Proactive information retrieval refers to a retrieval strategy that can extract relevant information implicitly without requiring explicit attention or interactions from the user (Li & Belkin, 2008). Primarily employed by modelled user’s search intent by observing the primary search and instantly retrieve information, without actively user to formulate queries (Dumais, Cutrell, Sarin, & Horvitz, 2004). Consequently, Intent estimation from limited primary search input is key challenge, for which traditional system asserts ‘Pre-search’ context (Kong, Zhang, Chang, & Allan, 2015) and user’s tasks (Saastamoinen, 2017), also serve twofold. First, to understand whether a user’s primary task, in contrast to just pre-search browsing history, can be used to predict search intentions, and second, to study if a user utilizes anticipated results and detect the tradeoffs it may offers (Koskela, Luukkonen, Ruotsalo 2018).

The Intent is a topical dimension of user search and characterizes ‘why’ the user is searching, and ‘how’ his search evolves during search progression (Marchionini, 2006). Characteristically, intent defined as ‘immediate reason, purpose, or goal’ that motivates a user to initiate or conduct a search (Nandi & Jagadish, 2011) and co-exist in three aspects, i.e. Pre-search, In-search and Pro-search of the user search context. A significant fraction of user searches is influenced by the user’s primary search aim i.e. ‘Pre-search’ context and others due to intermediate query or result in understanding ‘In-search’ context. An ideal information system would be able to predict the estimation of future intent (also known as ‘Pro-search’ context) based on the captured ‘Pre-search’ and ‘In-search’ context. The prediction of search context of futuristic search ‘Pro-search’ requires identification of co-relations between all three aspects of user search, therefore understanding ‘why’ users start searches and ‘how’ to predict search intent are multifaceted tasks (Van Rijsbergen, 1977; Daoud & Huang, 2013).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2019)
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing