Design Patterns for Fusion-Based Object Retrieval

TL;DR

This paper introduces two general design patterns, early and late fusion, for ranking objects without direct term representations by aggregating evidence from associated documents, applicable across various retrieval tasks.

Contribution

It proposes two reusable retrieval strategies, early and late fusion, that unify and encompass most existing object retrieval approaches.

Findings

Effective in expert finding, blog distillation, and vertical ranking

General patterns that unify previous retrieval methods

Demonstrated versatility across multiple object retrieval tasks

Abstract

We address the task of ranking objects (such as people, blogs, or verticals) that, unlike documents, do not have direct term-based representations. To be able to match them against keyword queries, evidence needs to be amassed from documents that are associated with the given object. We present two design patterns, i.e., general reusable retrieval strategies, which are able to encompass most existing approaches from the past. One strategy combines evidence on the term level (early fusion), while the other does it on the document level (late fusion). We demonstrate the generality of these patterns by applying them to three different object retrieval tasks: expert finding, blog distillation, and vertical ranking.