The Role-Relevance Model for Enhanced Semantic Targeting in Unstructured Text

TL;DR

This paper introduces a role-relevance model that improves personalized search by considering user roles, geographic and topical relevance, using an extended LDA method, resulting in a 20% precision boost.

Contribution

The paper presents a novel role-relevance algorithm that integrates geographic and topical relevance with standard keyword matching using an extended LDA for personalized search enhancement.

Findings

20% improvement in search precision over keyword search

Effective incorporation of user roles and geographic relevance

Novel extension to LDA for role-specific topical relevance

Abstract

Personalized search provides a potentially powerful tool, however, it is limited due to the large number of roles that a person has: parent, employee, consumer, etc. We present the role-relevance algorithm: a search technique that favors search results relevant to the user's current role. The role-relevance algorithm uses three factors to score documents: (1) the number of keywords each document contains; (2) each document's geographic relevance to the user's role (if applicable); and (3) each document's topical relevance to the user's role (if applicable). Topical relevance is assessed using a novel extension to Latent Dirichlet Allocation (LDA) that allows standard LDA to score document relevance to user-defined topics. Overall results on a pre-labeled corpus show an average improvement in search precision of approximately 20% compared to keyword search alone.