A Short Note on Proximity-based Scoring of Documents with Multiple Fields

TL;DR

This paper introduces a combined scoring method that integrates proximity considerations with multi-field document modeling, enhancing relevance ranking for complex documents.

Contribution

It proposes a novel scoring approach that merges BM25F and Expanded Span techniques for improved proximity-based document scoring.

Findings

Demonstrates improved relevance scoring for multi-field documents.

Provides a unified framework combining BM25F and Expanded Span.

Highlights potential for better retrieval performance in multi-field scenarios.

Abstract

The BM25 ranking function is one of the most well known query relevance document scoring functions and many variations of it are proposed. The BM25F function is one of its adaptations designed for modeling documents with multiple fields. The Expanded Span method extends a BM25-like function by taking into considerations of the proximity between term occurrences. In this note, we combine these two variations into one scoring method in view of proximity-based scoring of documents with multiple fields.