Sentence Sentiment Classification Using Convolutional Neural Network in Myanmar Texts

There are still few works on application of deep learning for Myanmar language. This paper presents an approach to use a convolutional neural network (CNN) model to classify sentence sentiment in Myanmar texts. A CNN model is constructed on the top of a word embedding model (i.e., Word2Vec), which converts words into vectors. The model classifies the input sentences and labels each sentence with positive, negative, neutral, unrelated and unreadable sentiments. The model is learnt from 1,152 sentences taken from the customers' reviews of products, which is provided by a telecommunication company. Then, the model is tested on 495 unseen sentences with the result of 86.26% accuracy and 82.58% average f-measure in prediction. The model is compared with the traditional machine learning (ML) classifiers, especially support vector machine (SVM), naïve Bayes (NB), and logistic regression (LR). The model outperforms these classifiers since SVM results in 64.44% accuracy, NB obtains 60.20% accuracy and LR gets 55.15% accuracy.

[1]  Mark Cieliebak,et al.  Sentiment Analysis using Convolutional Neural Networks with Multi-Task Training and Distant Supervision on Italian Tweets , 2019 .

[2]  Cícero Nogueira dos Santos,et al.  Deep Convolutional Neural Networks for Sentiment Analysis of Short Texts , 2014, COLING.

[3]  Yu Mon Aye,et al.  Senti-Lexicon and Analysis for Restaurant Reviews of Myanmar Text , 2018 .

[4]  Yoon Kim,et al.  Convolutional Neural Networks for Sentence Classification , 2014, EMNLP.

[5]  Pushpak Bhattacharyya,et al.  Sentiment Analysis : A Literature Survey , 2013, ArXiv.

[6]  M. Ali Akcayol,et al.  A comprehensive survey for sentiment analysis tasks using machine learning techniques , 2016, 2016 International Symposium on INnovations in Intelligent SysTems and Applications (INISTA).

[7]  Nadia Nedjah,et al.  Sentiment analysis using convolutional neural network via word embeddings , 2019, Evolutionary Intelligence.

[8]  Peerapon Vateekul,et al.  A study of sentiment analysis using deep learning techniques on Thai Twitter data , 2016, 2016 13th International Joint Conference on Computer Science and Software Engineering (JCSSE).

[9]  Keiron O'Shea,et al.  An Introduction to Convolutional Neural Networks , 2015, ArXiv.

[10]  Young-Seob Jeong,et al.  Sentiment Classification Using Convolutional Neural Networks , 2019, Applied Sciences.

[11]  Babar Hayat,et al.  Sentiment Analysis Using Deep Learning Techniques: A Review , 2017, International Journal of Advanced Computer Science and Applications.

[12]  Nadia Nedjah,et al.  Sentiment analysis using convolutional neural network with fastText embeddings , 2017, 2017 IEEE Latin American Conference on Computational Intelligence (LA-CCI).

[13]  Shuai Wang,et al.  Deep learning for sentiment analysis: A survey , 2018, WIREs Data Mining Knowl. Discov..

[14]  Kitsuchart Pasupa,et al.  Thai sentiment analysis with deep learning techniques: A comparative study based on word embedding, POS-tag, and sentic features , 2019, Sustainable Cities and Society.

[15]  Alessandro Moschitti,et al.  Twitter Sentiment Analysis with Deep Convolutional Neural Networks , 2015, SIGIR.