Sentiment Tagging with Partial Labels using Modular Architectures

TL;DR

This paper introduces a modular learning approach for sentiment analysis with partial labels, where separate modules are trained for sub-tasks and combined, improving learning efficiency and reducing supervision needs.

Contribution

The paper proposes a novel modular architecture for sequence prediction tasks with partial labels, enhancing learning constraints and supervision efficiency.

Findings

Modular approach improves sentiment analysis performance

Sharing information among modules constrains learning

Reduces supervision effort in training

Abstract

Many NLP learning tasks can be decomposed into several distinct sub-tasks, each associated with a partial label. In this paper we focus on a popular class of learning problems, sequence prediction applied to several sentiment analysis tasks, and suggest a modular learning approach in which different sub-tasks are learned using separate functional modules, combined to perform the final task while sharing information. Our experiments show this approach helps constrain the learning process and can alleviate some of the supervision efforts.