Applying skip-gram word estimation and SVM-based classification for opinion mining Vietnamese food places text reviews
Dang-Hung Phan, Tuan‐Dung Cao · 2014
In this paper, a framework for mining unstructured documents in the form of Vietnamese text comments about locations is proposed. In the first step, the evaluation of users in the form of text comments will be extracted to produce a set of sentences in Vietnamese standard grammars by web analytic processing. Through the second step, using Skip-gram based model, the similarity between each phrase in sentences will be detected. The core of this research focuses on contributing an approach for word representation that reflects Vietnamese semantic information contexts for analyzing as an input of a machine learning based classification, SVM. The application of this research is applied to the STAAR project's opinion mining system.