Double Embeddings and CNN-based Sequence Labeling for Aspect Extraction

TL;DR

This paper introduces a simple CNN model utilizing both general-purpose and domain-specific pre-trained embeddings for aspect extraction in sentiment analysis, achieving superior results without extra supervision.

Contribution

It presents the first CNN-based aspect extraction method using double embeddings, demonstrating improved performance over existing sophisticated models.

Findings

Outperforms state-of-the-art methods in aspect extraction

Uses dual embeddings without additional supervision

Achieves strong results with a simple CNN model

Abstract

One key task of fine-grained sentiment analysis of product reviews is to extract product aspects or features that users have expressed opinions on. This paper focuses on supervised aspect extraction using deep learning. Unlike other highly sophisticated supervised deep learning models, this paper proposes a novel and yet simple CNN model employing two types of pre-trained embeddings for aspect extraction: general-purpose embeddings and domain-specific embeddings. Without using any additional supervision, this model achieves surprisingly good results, outperforming state-of-the-art sophisticated existing methods. To our knowledge, this paper is the first to report such double embeddings based CNN model for aspect extraction and achieve very good results.