Opinion aspect extraction in Dutch childrens diary entries

TL;DR

This paper introduces a new Dutch children's diary dataset for aspect and opinion word extraction, applies deep learning models to improve aspect extraction, and achieves promising results in both review and children's language contexts.

Contribution

It creates a novel annotated Dutch children's dataset and adapts deep learning models for aspect extraction in this domain, improving upon previous results.

Findings

State-of-the-art results on Dutch review dataset

Promising aspect extraction results on children's data

Deep learning models effectively transfer across domains

Abstract

Aspect extraction can be used in dialogue systems to understand the topic of opinionated text. Expressing an empathetic reaction to an opinion can strengthen the bond between a human and, for example, a robot. The aim of this study is three-fold: 1. create a new annotated dataset for both aspect extraction and opinion words for Dutch childrens language, 2. acquire aspect extraction results for this task and 3. improve current results for aspect extraction in Dutch reviews. This was done by training a deep learning Gated Recurrent Unit (GRU) model, originally developed for an English review dataset, on Dutch restaurant review data to classify both opinion words and their respective aspects. We obtained state-of-the-art performance on the Dutch restaurant review dataset. Additionally, we acquired aspect extraction results for the Dutch childrens dataset. Since the model was trained on standardised language, these results are quite promising.