Shapley Additive Explanations for Text Classification and Sentiment Analysis of Internet Movie Database

被引：7

作者：

Dewi, Christine ^{[1
,2
]}

Tsai, Bing-Jun ^{[1
]}

Chen, Rung-Ching ^{[1
]}

机构：

[1] Chaoyang Univ Technol Taichung, Dept Informat Management, Taichung, Taiwan

[2] Satya Wacana Christian Univ, Fac Informat Technol, Salatiga, Indonesia

来源：

RECENT CHALLENGES IN INTELLIGENT INFORMATION AND DATABASE SYSTEMS, ACIIDS 2022 | 2022年 / 1716卷

关键词：

Natural language processing; Sentiment analysis; Shapley additive explanations (SHAP); Bidirectional encoder representations from transformers (BERT);

D O I：

10.1007/978-981-19-8234-7_6

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The application of Artificial Intelligence (AI) is increasing in areas like sentiment analysis and natural language processing (NLP). Automatic sentiment analysis provides a guide to capture the user emotions and classify the reviews into positive or negative. One of the challenges of using general lexicon analysis is its insensitivity to all domains. There arises a need for the interpretability of the output predicted from the AI sentiment analysis models. This paper developed a Shapley Additive Explanations for Text Classification (SHAP) based model to classify the user opinion texts into negative or positive labels. Our sentiment analysis model is evaluated on the Internet Movie Database (IMDB) datasets which have rich vocabulary and coherence of the textual data. Results showed that the model predicted 89% of the user reviews correctly. This model is very flexible for extending it to the unlabeled data.

引用

页码：69 / 80

页数：12

共 36 条

[1] Explaining individual predictions when features are dependent: More accurate approximations to Shapley values [J].

Aas, Kjersti ;

Jullum, Martin ;

Loland, Anders .

ARTIFICIAL INTELLIGENCE, 2021, 298

[2] A comprehensive survey of arabic sentiment analysis [J].

Al-Ayyoub, Mahmoud ;

Khamaiseh, Abed Allah ;

Jararweh, Yaser ;

Al-Kabi, Mohammed N. .

INFORMATION PROCESSING & MANAGEMENT, 2019, 56 (02) :320-342

[3] Lexicon based feature extraction for emotion text classification [J].

Bandhakavi, Anil ;

Wiratunga, Nirmalie ;

Padmanabhan, Deepak ;

Massie, Stewart .

PATTERN RECOGNITION LETTERS, 2017, 93 :133-142

[4] JUMRv1: A Sentiment Analysis Dataset for Movie Recommendation [J].

Chatterjee, Shuvamoy ;

Chakrabarti, Kushal ;

Garain, Avishek ;

Schwenker, Friedhelm ;

Sarkar, Ram .

APPLIED SCIENCES-BASEL, 2021, 11 (20)

[5]

Chen R.-C., 2020, Int. J. Appl. Sci. Eng., V17, P237

[6] Selecting critical features for data classification based on machine learning methods [J].

Chen, Rung-Ching ;

Dewi, Christine ;

Huang, Su-Wen ;

Caraka, Rezzy Eko .

JOURNAL OF BIG DATA, 2020, 7 (01)

[7] Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN [J].

Chen, Tao ;

Xu, Ruifeng ;

He, Yulan ;

Wang, Xuan .

EXPERT SYSTEMS WITH APPLICATIONS, 2017, 72 :221-230

[8]

Chiny M, 2021, INT J ADV COMPUT SC, V12, P265

[9] Prediction of follower jumps in cam-follower mechanisms: The benefit of using physics-inspired features in recurrent neural networks [J].

De Groote, Wannes ;

Van Hoecke, Sofie ;

Crevecoeur, Guillaume .

MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2022, 166

[10]

Devlin J, 2019, 2019 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL HLT 2019), VOL. 1, P4171

← 1 2 3 4 →