BALAS: Empirical Bayesian learning in the relevance feedback for image retrieval

Abstract

This paper is on user relevance feedback in image retrieval. We take this problem as a standard two-class pattern classification problem aiming at refining the retrieval precision by learning through the user relevance feedback data. However, we have investigated the problem by noting two important unique characteristics of the problem: small sample collection and asymmetric sample distributions between positive and negative samples. We have developed a novel approach to empirical Bayesian learning to solve for this problem by explicitly exploiting the two unique characteristics, which is the methodology of BAyesian Learning in Asymmetric and Small sample collections, thus called BALAS. In BALAS different learning strategies are used for positive and negative sample collections, respectively, based on the two unique characteristics. By defining the relevancy confidence as the relevant posterior probability, we have developed an integrated ranking scheme in BALAS which complementarily combines the subjective relevancy confidence and the objective similarity measure to capture the overall retrieval semantics. The experimental evaluations have confirmed the rationale of the proposed ranking scheme, and have also demonstrated that BALAS is superior to an existing relevance feedback method in the current literature in capturing the overall retrieval semantics.

Second Generation of Mass Estimation

“…Anomaly detection aims to rank anomalous points higher than normal points; information retrieval aims to rank points similar to a query higher than dissimilar points. Many existing methods (e.g., ; Zhang and Zhang 2006) have employed density to provide the ranking; but density estimation is not designed to provide a ranking.…”

“…Anomaly detection aims to rank anomalous points higher than normal points; information retrieval aims to rank points similar to a query higher than dissimilar points. Many existing methods (e.g., Zhang and Zhang 2006) have employed density to provide the ranking; but density estimation is not designed to provide a ranking.We show in this paper that a new base modelling mechanism called mass estimation possesses different properties from those offered by density estimation:• A mass distribution stipulates an ordering from core points to fringe points in a data cloud. In addition, this ordering accentuates the fringe points with a concave function derived from data, resulting in fringe points having markedly smaller mass than points close to the core points.…”

Relevance feature mapping for content-based image retrieval

“…Third, some methods transform the CBIR problem into a classification problem, and solve it using a classification technique such as support vector machine [11], and Bayesian method [17]. A representative method called BALAS [17] first estimates the probability density function of positive class as well as negative ones, and then produces the ranking using a Bayesian learning framework.…”

“…A representative method called BALAS [17] first estimates the probability density function of positive class as well as negative ones, and then produces the ranking using a Bayesian learning framework. However, most classification methods are designed to classify images into a fixed number of classes and are not designed for ranking images.…”

“…Many existing image retrieval systems index and search the images based on textual information such as keywords, surrounding text, etc.. However, the text-based search suffers from the following inherent drawbacks [17,4]: (i) the textual information is usually nonexistent or incomplete with the emergence of massive image databases; (ii) the textual description is not sufficient for depicting subjective semantics since different people may describe the content of 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. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.…”