Recursive Neural Language Architecture for Tag Prediction

TL;DR

This paper introduces a neural feedback relevance model that learns weighted tag representations from content, significantly improving tag recommendation quality over existing methods.

Contribution

It proposes a novel neural model that captures weighted feature representations for tags, overcoming linear compositional limitations of previous models.

Findings

Significant improvement in recommendation quality on two datasets

Outperforms baseline models in tag prediction accuracy

Demonstrates effectiveness of weighted feature representations

Abstract

We consider the problem of learning distributed representations for tags from their associated content for the task of tag recommendation. Considering tagging information is usually very sparse, effective learning from content and tag association is very crucial and challenging task. Recently, various neural representation learning models such as WSABIE and its variants show promising performance, mainly due to compact feature representations learned in a semantic space. However, their capacity is limited by a linear compositional approach for representing tags as sum of equal parts and hurt their performance. In this work, we propose a neural feedback relevance model for learning tag representations with weighted feature representations. Our experiments on two widely used datasets show significant improvement for quality of recommendations over various baselines.