Supervised Complementary Entity Recognition with Augmented Key-value Pairs of Knowledge

TL;DR

This paper introduces a supervised method for recognizing complementary entities in reviews by augmenting a CRF model with automatically learned domain knowledge, improving extraction accuracy.

Contribution

It proposes a novel knowledge-based CRF model that automatically expands domain knowledge from unlabeled data to enhance complementary entity recognition.

Findings

KCRF outperforms baseline models in CER tasks.

Knowledge augmentation improves recognition accuracy.

Effective use of unlabeled reviews for knowledge expansion.

Abstract

Extracting opinion targets is an important task in sentiment analysis on product reviews and complementary entities (products) are one important type of opinion targets that may work together with the reviewed product. In this paper, we address the problem of Complementary Entity Recognition (CER) as a supervised sequence labeling with the capability of expanding domain knowledge as key-value pairs from unlabeled reviews, by automatically learning and enhancing knowledge-based features. We use Conditional Random Field (CRF) as the base learner and augment CRF with knowledge-based features (called the Knowledge-based CRF or KCRF for short). We conduct experiments to show that KCRF effectively improves the performance of supervised CER task.