Abstract:Document-level review sentiment classification aims to predict the sentiment category for given review documents written by users for products. Most of the existing methods focus on generating a good review document representation and classifying the review document directly. However, on the one hand, as document-level review sentiment classification usually includes many sentiment categories and the difference between these sentiment categories is not obvious, it may be difficult to obtain satisfying result by direct classification. On the other hand, this once classification process with review representation may fail to well interpret how the results are achieved. In addition, although some information such as user preference and product characteristics are incorporated when building models, the interactive information between user and product are usually ignored. In this paper, inspired by the deductive reasoning strategy of human doing multiple choice questions, we are motivated to propose a Two-Stage Sentiment Classification (TSSC) model to classify review documents in two stages: (1) Coarse classification stage, where model mainly adopts user-product interactive information to pre-judge the sentiment tendency of the review document without considering the review information; (2) Fine classification stage, where model uses text information of the review document for further classification based on the sentiment tendency obtained in coarse classification stage. Finally, the sentiment classification task is accomplished by combining both the results of coarse classification and fine classification. The experimental results demonstrate that our TSSC model significantly outperforms most of the related models (e.g., Trigram and NSC+UPA) on IMDB and Yelp datasets in terms of classification accuracy. When compared with the state-of-the-art HUAPA model, our TSSC model not only achieves slightly more accurate performance, but also has lower time complexity and stronger interpretability.

Spectral Clustering-Based Semi-supervised Sentiment Classification.

Semi-supervised clustering based on spectral clustering

Semi-supervised Sentiment Classification Using Ranked Opinion Words

Sentiment Classification for Chinese Reviews: a Comparison Between SVM and Semantic Approaches

Sentiment Classification Using Subjective and Objective Views

Cooperative Hybrid Semi-Supervised Learning For Text Sentiment Classification

A Mixture Language Model for the Classification of Chinese Online Reviews.

Semi-supervised Learning with Generative Model for Sentiment Classification of Stock Messages

Spectral Transformation Approaches to Semi-supervised Learning

Sentiment Classification Using Supervised and Semi-Supervised Conditional Maximum Entropy Modeling.

Self-supervised spectral clustering with spectral embedding for hyperspectral image classification

Application of a clustering method on sentiment analysis

Comparison of semi-supervised and supervised approaches for classification of e-nose datasets: Case studies of tomato juices

Exploiting Co-occurrence Opinion Words for Semi-supervised Sentiment Classification.

Target-oriented Sentiment Classification with Sequential Cross-modal Semantic Graph

Affective Classification In Video Based On Semi-Supervised Learning

Semantic Spectral Clustering with Contrastive Learning and Neighbor Mining

Semi-supervised Target-Oriented Sentiment Classification

Semi-supervised Sentiment Classification Based on Auxiliary Task Learning.

Two-stage Sentiment Classification Based on User-Product Interactive Information

A Multi-Granularity Fuzzy Computing Model for Sentiment Classification of Chinese Reviews