Extending Probabilistic Data Fusion Using Sliding Windows

TL;DR

This paper introduces SlideFuse, a probabilistic data fusion method using sliding windows to improve relevance estimation when limited data is available, outperforming existing techniques like CombMNZ, ProbFuse, and SegFuse.

Contribution

SlideFuse extends probabilistic data fusion by incorporating sliding windows to enhance relevance estimation with sparse data, advancing the state-of-the-art.

Findings

SlideFuse outperforms CombMNZ, ProbFuse, and SegFuse in relevance estimation.

Sliding windows improve data fusion accuracy with limited relevance information.

SlideFuse demonstrates superior ranking performance in experiments.

Abstract

Recent developments in the field of data fusion have seen a focus on techniques that use training queries to estimate the probability that various documents are relevant to a given query and use that information to assign scores to those documents on which they are subsequently ranked. This paper introduces SlideFuse, which builds on these techniques, introducing a sliding window in order to compensate for situations where little relevance information is available to aid in the estimation of probabilities. SlideFuse is shown to perform favourably in comparison with CombMNZ, ProbFuse and SegFuse. CombMNZ is the standard baseline technique against which data fusion algorithms are compared whereas ProbFuse and SegFuse represent the state-of-the-art for probabilistic data fusion methods.