Representation Learning Models for Entity Search

TL;DR

This paper introduces new representation learning models for entity search that generate low-dimensional vectors for queries, entities, and descriptions, improving search accuracy across various entity types.

Contribution

The paper proposes three novel learning strategies for representing entities and queries as vectors, analyzing their strengths and weaknesses, and demonstrating superior performance over existing methods.

Findings

Outperforms keyword matching and vanilla word2vec models

Adapts well to different entity types like movies and restaurants

Fast training and easy extension to similar tasks

Abstract

We focus on the problem of learning distributed representations for entity search queries, named entities, and their short descriptions. With our representation learning models, the entity search query, named entity and description can be represented as low-dimensional vectors. Our goal is to develop a simple but effective model that can make the distributed representations of query related entities similar to the query in the vector space. Hence, we propose three kinds of learning strategies, and the difference between them mainly lies in how to deal with the relationship between an entity and its description. We analyze the strengths and weaknesses of each learning strategy and validate our methods on public datasets which contain four kinds of named entities, i.e., movies, TV shows, restaurants and celebrities. The experimental results indicate that our proposed methods can adapt to different types of entity search queries, and outperform the current state-of-the-art methods based on keyword matching and vanilla word2vec models. Besides, the proposed methods can be trained fast and be easily extended to other similar tasks.