Abstract:Pseudo-relevance feedback (PRF) has evident potential for enriching the representation of short queries. Traditional PRF methods treat top-ranked documents as feedback, since they are assumed to be relevant to the query. However, some of these feedback documents may actually distract from the query topic for a range of reasons and accordingly downgrade PRF system performance. Such documents constitute negative examples (negative feedback) but could also be valuable in retrieval. In this paper, a novel framework of query language model construction is proposed in order to improve retrieval performance by integrating both positive and negative feedback. First, an improvement-based method is proposed to automatically identify the types of feedback documents (i.e. positive or negative) according to whether the document enhances the retrieval's effectiveness. Subsequently, based on the learned positive and negative examples, the positive feedback models and the negative feedback models are estimated using an Expectation-Maximization algorithm with the assumptions: the positive term distribution is affected by the context term distribution and the negative term distribution is affected by both the positive term distribution and the context term distribution (such that the positive feedback model upgrades the rankings of relevant documents and the negative feedback model prunes the irrelevant documents from a query). Finally, a content-based representativeness criterion is proposed in order to obtain the representative negative feedback documents. Experiments conducted on the TREC collections demonstrate that our proposed approach results in better retrieval accuracy and robustness than baseline methods. (C) 2018 Elsevier Ltd. All rights reserved.

Relevance feedback based query optimization

Fully utilize feedbacks: language model based relevance feedback in information retrieval

A Powerful Relevance Feedback Mechanism for Content-Based 3D Model Retrieval

The Study of Methods for Language Model Based Positive and Negative Relevance Feedback in Information Retrieval

A Multiple Relevance Feedback Strategy with Positive and Negative Models.

A Document Relevance Based Search Result Re-Ranking

A Survey on Relevance Feedback Techniques in Content-based Medical Image Retrieval

A Statistical Model for Relevance Feedback in Information Retrieval

Towards Optimal Query Design For Relevance Feedback In Image Retrieval

A Relevance Feedback System for CBIR with Long-Term Learning

Research on Model and Method of Relevance Feedback Mechanism in Image Retrieval.

Modeling Positive and Negative Feedback for Improving Document Retrieval.

Relevance Feedback in Content-Based Image Retrieval: A Survey

Evaluation of relevance feedback methods for 3D model retrieval

Relevance Feedback Based on Query Refining and Feature Database Updating in CBIR System

Relevance Feedback Technologies in Search Engines

Relevance Feedback For Keyword And Visual Feature-Based Image Retrieval

A GA-based query optimization method for web information retrieval

Weakly Supervised Relevance Feedback Based on an Improved Language Model

A Powerful Partial Relevance Feedback for 3D Model Retrieval

Relevance Feedback Based on Constrained Random Seiection and Retrieval Set