Abstract
The accuracy of online review mining for e-commerce products is of great value to customer and product matching portrait. Mining the fine-grained aspect in reviews is a key indicator. It can better analyze the emotion tendency of online reviews and understand the advantages and disadvantages of evaluation objects. In this paper, we propose a semi-supervised learning method to extract product aspects and description of aspects. Specifically, we firstly construct word vector space model of large scale reviews with deep learning, then get the list of similar words based on the model. Finally, the fine-grained aspect sets are obtained by classification algorithm. The results of the study show that the efficiency of fine-grained extraction is improved by using semi-supervised method.
Recommended Citation
Xia, Huosong and Yang, Yitai, "Fine-grained Aspect Extraction for Online Reviews of E-commerce Products Based on Semi-supervised Learning" (2018). WHICEB 2018 Proceedings. 36.
https://aisel.aisnet.org/whiceb2018/36