Mining, Ranking and Recommending Entity Aspects

Abstract

Entity queries constitute a large fraction of web search queries and most of these queries are in the form of an entity mention plus some context terms that represent an intent in the context of that entity. We refer to these entity-oriented search intents as entity aspects. Recognizing entity aspects in a query can improve various search applications such as providing direct answers, diversifying search results, and recommending queries. In this paper we focus on the tasks of identifying, ranking, and recommending entity aspects, and propose an approach that mines, clusters, and ranks such aspects from query logs. We perform large-scale experiments based on users' search sessions from actual query logs to evaluate the aspect ranking and recommendation tasks. In the aspect ranking task, we aim to satisfy most users' entity queries, and evaluate this task in a queryindependent fashion. We find that entropy-based methods achieve the best performance compared to maximum likelihood and language modeling approaches. In the aspect recommendation task, we recommend other aspects related to the aspect currently being queried. We propose two approaches based on semantic relatedness and aspect transitions within user sessions and find that a combined approach gives the best performance. As an additional experiment, we utilize entity aspects for actual query recommendation and find that our approach improves the effectiveness of query recommendations built on top of the query-flow graph.

Multiple Models for Recommending Temporal Aspects of Entities

“…[26] focuses on salient ranking features in microblogs. Reinanda et al [22] start from the task of mining entity aspects in the query logs, then propose salience-favor methods for ranking and recommending these aspects. When regarding an aspect as an entity, related work connected to temporal IR is [31], where they study the task of time-aware entity recommendation using a probabilistic approach.…”

“…We will approach the task of recommending temporal entity aspect as a ranking task. We first define the notions of an entity query, a temporal entity aspect, developed from the definition of entity aspect in [22], and an event entity . We then formulate the task of recommending temporal entity aspects.…”

“…An entity query q e is a query that is represented by one Wikipedia entity e. We consider q e as the representation of e. Definition 2. Given a "search task" defined as an atomic information need, a temporal "entity aspect" is an entity-oriented search task with time-aware intent.. An entity-oriented search task is a set of queries that represent a common task in the context of an entity, grouped together if they have the same intent [22]. We will use the notion of query q to indicate an entity aspect a interchangeably hereafter.…”

Towards an Understanding of Entity-Oriented Search Intents

“…Search intents have been studied in previous work. Reinanda et al [12] explore entity aspects in user interaction log data. Beyond finding aspects by comparing clustering methods over refiners, they address the tasks of ranking the intents for a given entity independently from a query and recommending aspects.…”

IntentsKB

“…Most related to our paper is the work by Reinanda et al [9], who explore entity aspects in user interaction log data. Beyond finding aspects by comparing clustering methods over refiners, they address the tasks of ranking such intents for a given entity independently from a query and recommending aspects.…”