Aspect-Level Sentiment Analysis Based on Position Features Using Multilevel Interactive Bidirectional GRU and Attention Mechanism
Open Access
- 8 July 2020
- journal article
- research article
- Published by Wiley in Discrete Dynamics in Nature and Society
- Vol. 2020, 1-13
Abstract
The aim of aspect-level sentiment analysis is to identify the sentiment polarity of a given target term in sentences. Existing neural network models provide a useful account of how to judge the polarity. However, context relative position information for the target terms is adversely ignored under the limitation of training datasets. Considering position features between words into the models can improve the accuracy of sentiment classification. Hence, this study proposes an improved classification model by combining multilevel interactive bidirectional Gated Recurrent Unit (GRU), attention mechanisms, and position features (MI-biGRU). Firstly, the position features of words in a sentence are initialized to enrich word embedding. Secondly, the approach extracts the features of target terms and context by using a well-constructed multilevel interactive bidirectional neural network. Thirdly, an attention mechanism is introduced so that the model can pay greater attention to those words that are important for sentiment analysis. Finally, four classic sentiment classification datasets are used to deal with aspect-level tasks. Experimental results indicate that there is a correlation between the multilevel interactive attention network and the position features. MI-biGRU can obviously improve the performance of classification.Keywords
Funding Information
- Ministry of Education of the People's Republic of China (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- National Natural Science Foundation of China (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Science and Technology Department of Sichuan Province (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Science and Technology Department of Sichuan Province (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Science and Technology Department of Sichuan Province (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Science and Technology Department of Sichuan Province (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Science and Technology Department of Sichuan Province (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Chengdu Science and Technology Bureau (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Fund of Sichuan Educational Committee (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Fund of Sichuan Educational Committee (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Foundation of Cyberspace Security Key Laboratory of Sichuan Higher Education Institutions (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
- Xihua University (Z2015100, 61902324, 2016JY0244, 2017JQ0059, 2019GFW131, 2020JY, 2020GFW, 2017-RK00-00026-ZF, 15ZB0134, 17ZA0360, sjzz2016-73)
This publication has 16 references indexed in Scilit:
- Feature selection and ensemble construction: A two-step method for aspect based sentiment analysisKnowledge-Based Systems, 2017
- A Hierarchical Model of Reviews for Aspect-based Sentiment Analysis2016
- Attention-based LSTM for Aspect-level Sentiment Classification2016
- Aspect Level Sentiment Classification with Deep Memory Network2016
- Survey on Aspect-Level Sentiment AnalysisIEEE Transactions on Knowledge and Data Engineering, 2015
- Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks2015
- DCU: Aspect-based Polarity Classification for SemEval Task 42014
- Glove: Global Vectors for Word Representation2014
- Survey on mining subjective data on the webData Mining and Knowledge Discovery, 2011
- Bidirectional recurrent neural networksIEEE Transactions on Signal Processing, 1997