Sentiment Analysis in Scholarly Book Reviews

TL;DR

This paper introduces a supervised sentiment analysis method for scholarly book reviews, combining features to identify opinion targets and polarities, and demonstrates robustness across French and English domains.

Contribution

It presents a novel approach for sentiment analysis in scholarly reviews, including a new labeled corpus and cross-domain evaluation.

Findings

Methods are robust for English restaurant reviews.

Methods perform well on French scholarly book reviews.

Cross-domain and cross-language robustness demonstrated.

Abstract

So far different studies have tackled the sentiment analysis in several domains such as restaurant and movie reviews. But, this problem has not been studied in scholarly book reviews which is different in terms of review style and size. In this paper, we propose to combine different features in order to be presented to a supervised classifiers which extract the opinion target expressions and detect their polarities in scholarly book reviews. We construct a labeled corpus for training and evaluating our methods in French book reviews. We also evaluate them on English restaurant reviews in order to measure their robustness across the domains and languages. The evaluation shows that our methods are enough robust for English restaurant reviews and French book reviews.