Manifold-based Sparse Representation for Opinion Mining

Abstract

Abstract This paper focuses on how to enhance feature representation for opinion mining. The classical feature representation methods suffer from high dimensionality, sparsity , noisy, irrelevant and redundant information. It is proposed to exploit the manifold assumption and sparse property as prior knowledge for opinion representation to learn effective features. First, the graph representation of user reviews based on the mentioned prior knowledge is learned. Then, the spectral properties of the learned graph are exploited to present data in a new feature space. The proposed algorithm is applied to four various common input features on two benchmark datasets, Internet Movie Database (IMDB) and Amazon review dataset. Our experiments reveal that the proposed algorithm yields considerable enhancements in terms of F-measure, accuracy, and other standard performance measures compared to the combination of state-of-the-art features with various classifiers. The highest classification accuracies of 99.15 and 91.97 are obtained in the proposed method on IMDB and Amazon exploiting linear SVM classifier, respectively. The impact of parameters of the proposed algorithm is also investigated in this paper.