Massive Query Expansion by Exploiting Graph Knowledge Bases for Image Retrieval

Abstract

Annotation-based techniques for image retrieval suffer from sparse and short image textual descriptions. Moreover, users are often not able to describe their needs with the most appropriate keywords. This situation is a breeding ground for a vocabulary mismatch problem resulting in poor results in terms of retrieval precision. In this paper, we propose a query expansion technique for queries expressed as keywords and short natural language descriptions. We present a new massive query expansion strategy that enriches queries using a graph knowledge base by identifying the query concepts, and adding relevant synonyms and semantically related terms. We propose a topological graph enrichment technique that analyzes the network of relations among the concepts, and suggests semantically related terms by path and community detection analysis of the knowledge graph. We perform our expansions by using two versions of Wikipedia as knowledge base achieving improvements of the system's precision up to more than 27%.

Query expansion using Wikidata attributes’ values

“…the click activity of the users. The information in the logs of ancient queries that may be used to expand the user"s query is the relationship between queries and selected documents [12]. In [13] the authors extract probabilistic correlations between query terms and document terms, by analyzing query logs, in order to determine expansion terms.…”

“…In [13] the authors extract probabilistic correlations between query terms and document terms, by analyzing query logs, in order to determine expansion terms. Yet, this technique is not efficient for systems that do not have large logs [12]. In the last decades, the IR field has also known an integration of the context aspect to query expansion.…”

Put Three and Three Together

“…For instance, in social networks, communities identify groups of users with similar interests, locations, friends or occupations. This information is useful to perform more focused marketing campaigns [Wang et al 2009], to craft new visual representations of data [Di Giacomo et al 2007], increasing data locality thanks to a more coalesced data placement [Prat-Pérez et al 2011], finding expansion terms in query engines [Guisado-Gámez et al 2014] or for item suggestion [Sozio and Gionis 2010].…”

Mining document, concept, and term associations for effective biomedical retrieval: introducing MeSH-enhanced retrieval models

“…User queries have been found to be short, generally consisting of one or two words on average (Zeng et al 2002), and their terms are significantly different from the ones in professional thesauri (Zhang et al 2008) and in documents. One reason for that is the inexperience with topics (Guisado-Gá mez et al 2013). If users are not familiar with the topic, they may have difficulties in formulating effective queries (Zeng et al 2004).…”