Role-explicit query identification and intent role annotation

Abstract

Understanding the information need or intent encoded within a query has long been regarded as an essential factor of effective information retrieval. For better query representation and understanding, two intent roles (kernel-object and modifier) are introduced to structurally parse a class of role-explicit queries, which constitute a majority of common user queries. Furthermore, we focus on two research problems: RP-1: Given a role-explicit query, how to identify the kernel-object and modifier, namely intent role annotation; RP-2: How to determine whether an arbitrary query is role-explicit or not. To solve RP-1, we propose a simplified word n-gram role model (SWNR), which quantifies the generating probability of a role-explicit query and performs intent role annotation effectively. Using a set of discriminative features, we build classifiers to address RP-2 in a supervised manner. The experimental results show that: (1) SWNR can achieve a satisfactory performance, more than 73% in terms of different metrics; (2) The classifiers can achieve more than 90% precision in identifying role-explicit queries; (3) Compared with traditional techniques for query representation and understanding, e.g., name entity recognition in query and class-level query intent inference, intent role annotation provides a more flexible framework and a number of applications can benefit from annotating role-explicit queries, such as intent mining and diversified document ranking.

Automatic role‐explicit query extraction: a divide‐and‐conquer system leveraging on users' reformulating behaviors

“…Instead of performing a fine‐grained classification of entities in queries like NERQ or naïve query segmentation, annotating user queries with intent roles provides a more flexible framework for query representation and understanding [9]. Though the techniques in the previous works cannot be directly used to perform role‐explicit query extraction, they give us insights of how to better capture the features of user queries.…”

“…This strategy suffers from some serious limitations on the feasibility. The current solution for role‐explicit query identification and intent role annotation proposed by Yu and Ren [9] requires a corpus of sufficient role‐explicit queries as the training corpus, which will result in a circular‐dependency problem when used in our case. Independent query segmentation in the second step is potentially ambiguous.…”

“…Analogous to the semantic role labeling in natural language processing, to get a better query representation and understanding, Yu and Ren proposed the scheme of intent role annotation [9]. Two intent roles kernel‐object and modifier are introduced to structurally parse a class of role‐explicit queries, which constitute a majority of common user queries.…”

“…Our work essentially is offline query log mining, and its value lies in the following: A corpus of sufficient annotated role‐explicit queries is a valuable resource for further knowledge mining. For example, it can be used as the training corpus for classifiers of role‐explicit query identification and intent role annotation [9]. Different modifiers aggregated from different users for the same kernel‐object are strong indicators of diverse kernel‐object‐oriented intents, which are of great importance for query recommendation [10], intent mining, and document ranking [11].…”

Search Result Diversification via Filling Up Multiple Knapsacks

“…Instead of formulating natural language queries, the vast majority of users are submitting short queries with little or no context, which are often ambiguous and/or underspecified [32]. Consider the well-worn query Harry Potter for example, it could refer Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page.…”