Representation Learning Models for Entity Search

• DOI: 10.48550/arxiv.1610.09091
• Publication Date: 2016-01-01
• Authors: E Shijia; Yang Xiang; Mohan Zhang
• Keywords: Representation

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