Advertisement

Multimedia Tools and Applications

, Volume 78, Issue 22, pp 32357–32378 | Cite as

Opinion mining in Persian language using a hybrid feature extraction approach based on convolutional neural network

  • Shima Zobeidi
  • Marjan NaderanEmail author
  • Seyyed Enayatallah Alavi
Article
  • 39 Downloads

Abstract

Nowadays, huge amounts of text data are generated due to the increase of communications, over various web sites and applications. Evaluation and extraction of information from these data is an important task, one way of which is named opinion mining. The purpose of this paper is sentiment analysis of users’ opinions about various products. The proposed system classifies opinions at the sentence level based on emotions into two and multiple classes by deep learning methods. To this end, three main phases are taken: the first step contains sentences preparation for the input matrix which itself is accomplished in two levels: word-level and character-level. In word-level, each word in each sentence is given to the word2vec algorithm. In character-level, for each character in each sentence, the proposed method computes a numerical vector and creates a matrix. Next, the feature extraction part is executed which includes a Convolutional Neural Network (CNN). The generated matrices in the previous levels for each sentence are given to the CNN for embedding each sentence and therefore, utilizing both word2vec and CNN for extracting features. In the final step, the generated vectors are given to the Bidirectional Long Short Term Memory (Bi-LSTM) network for sentiment classification, not used in any of the previous methods. The performance of the proposed algorithm has been investigated on the Digikala Persian dataset on mobile and digital cameras. Results show that the proposed algorithm reaches an accuracy of 95% for two classes and 92% for multi-class classification which is comparable with previous algorithms.

Keywords

Deep learning Text mining Opinion mining Convolutional neural network Bi-LSTM Word2vec Character-level Sentiment analysis 

Notes

Acknowledgements

The authors would like to thank Shahid Chamran University of Ahvaz High PerformanceComputing Center (SCU-HPCC) for providing computing resources for this project.

References

  1. 1.
    Alshari EM, Azman A, Doraisamy S, Mustapha N, Alkeshr M (2017) Improvement of sentiment analysis based on clustering of Word2vec features. In: 28th international workshop on database and expert systems applications (DEXA)Google Scholar
  2. 2.
    Amiri F, Scerri S, Khodashahi M, Iais F, Augustin S (2015) Lexicon-based sentiment analysis for Persian text, pp 9–16Google Scholar
  3. 3.
    An Y, Tang X, Xie B (2017) Sentiment analysis for short Chinese text based on character-level methods. In: 9th international conference on knowledge and smart technology (KST)Google Scholar
  4. 4.
    Anwar S, Hwang K, sung W (2015) Fixed point optimization of deep convolution neural networks of object recognition. ICASSP, pp 1131–1135Google Scholar
  5. 5.
    Bagheri B, Saraee M, de Jong F (2013) Sentiment classification in Persian: introducing a mutual information-based method for feature selection. In: 2013 21st Iranian conference on electrical engineering (ICEE), pp 1–6Google Scholar
  6. 6.
    Basiri ME, Kabiri A (2017) Sentence-level sentiment analysis in Persian. 3rd International Conference on Pattern Recognition and Image Analysis (IPRIA), Shahrekord, Iran, pp 84–89Google Scholar
  7. 7.
    Basiri M, Nilchi A, Ghassem-Aghaee N (2014) A framework for sentiment analysis in Persian. Open Transactions on Information Processing 1:1–14CrossRefGoogle Scholar
  8. 8.
    Basiri M, Naghsh-Nilchi A, Ghassem-Aghaee N (2014) A framework for sentiment analysis in persian. Open Trans Inf Process 1(3):1–14Google Scholar
  9. 9.
    Boudad N, Faizi R, Oulad Haj Thami R, Chiheb R (2017) Sentiment analysis in Arabic: a review of the literature. Ain Shams Engineering Journal 9:2479–2490CrossRefGoogle Scholar
  10. 10.
    Chau N, Phan V, Nguyen M (2016) Deep learning and sub-tree mining for document level sentiment classification. In: Eighth International Conference on Knowledge and Systems Engineering (KSE)Google Scholar
  11. 11.
    De Mulder W, Bethard S, Moens M (2015) A survey on the application of recurrent neural networks to statistical language modeling. Comput Speech Lang 30(1):61–98CrossRefGoogle Scholar
  12. 12.
    Dhomne A, Sa P (2018) Face verification using deep learning. JIMS8I International Journal of Information Communication and Computing Technology 6(1):332CrossRefGoogle Scholar
  13. 13.
    Digikala (2017) [Online]. Available: http://www.digikala.com. Accessed: 15 Feb 2017
  14. 14.
    Elman J (1990) Finding structure in time. Cogn Sci 14(2):179–211CrossRefGoogle Scholar
  15. 15.
    Enríquez F, Troyano JA, López-Solaz T (2016) An approach to the use of word embeddings in an opinion classification task. Expert Syst Appl 66:1–6CrossRefGoogle Scholar
  16. 16.
    Gao Y, Rong W, Shen Y, Xiong Z (2016) Convolutional neural network based sentiment analysis using Adaboost combination. In: International Joint Conference on Neural Networks (IJCNN)Google Scholar
  17. 17.
    Gopalakrishnan V, Ramaswamy C (2017) Patient opinion mining to analyze drugs satisfaction using supervised learning. Journal of Applied Research and Technology 15(4):311–319CrossRefGoogle Scholar
  18. 18.
    Graves A (2013) Generating sequences with recurrent neural networks. arXiv preprint arXiv:1308.0850Google Scholar
  19. 19.
    Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780CrossRefGoogle Scholar
  20. 20.
    Krouska A, Troussas C, Virvou M (2016) The effect of preprocessing techniques on twitter sentiment analysis. In: 7th International Conference on Information, Intelligence, Systems & Applications (IISA)Google Scholar
  21. 21.
    Lang K, Waibel A, Hinton G (1990) A time-delay neural network architecture for isolated word recognition. Neural Netw 3(1):23–43CrossRefGoogle Scholar
  22. 22.
    Lau R, Xia Y, Ye Y (2014) A probabilistic generative model for mining cybercriminal networks from online social media. IEEE Comput Intell Mag 9(1):31–43CrossRefGoogle Scholar
  23. 23.
    Le QV, Mikolov, T (2014) Distributed representations of sentences and documents, computation and languageGoogle Scholar
  24. 24.
    Lei X, Qian X, Zhao G (2016) Rating prediction based on social sentiment from textual reviews. IEEE Transactions on Multimedia 18(9):1910–1921CrossRefGoogle Scholar
  25. 25.
    Li X, Xie H, Chen L, Wang J, Deng X (2014) News impact on stock price return via sentiment analysis. Knowl-Based Syst 69:14–23CrossRefGoogle Scholar
  26. 26.
    Li H, Peng Q, Guan X (2016) Sentence level opinion mining of hotel comment. IEEE International Conference on Information and Automation (ICIA)Google Scholar
  27. 27.
    Li Q, Jin Z, Wang C, Zeng D (2016) Mining opinion summarizations using convolutional neural networks in Chinese microblogging systems. Knowl-Based Syst 107:289–300CrossRefGoogle Scholar
  28. 28.
    Li X, Peng L, Yao X, Cui S, Hu Y, You C, Chi T (2017) Long short-term memory neural network for air pollutant concentration predictions: method development and evaluation. Environ Pollut 231:997–1004CrossRefGoogle Scholar
  29. 29.
    Liu B (2012) Sentiment analysis and opinion mining. Morgan & Claypool, San RafaelCrossRefGoogle Scholar
  30. 30.
    Liu Z, Chen Y, Tang B, Wang X, Chen Q, Li H, Wang J, Deng Q, Zhu S (2015) Automatic de-identification of electronic medical records using token-level and character-level conditional random fields. J Biomed Inform 58:S47–S52CrossRefGoogle Scholar
  31. 31.
    McLeod P, Shallice T, Plaut D (2000) Attractor dynamics in word recognition: converging evidence from errors by normal subjects, dyslexic patients and a connectionist model. Cognition 74(1):91–114CrossRefGoogle Scholar
  32. 32.
    Mikolov T, Sutskever I, Chen K, Corrado GS, Dean J (2013) Distributed representations of words and phrases and their compositionality. In: Advances in neural information processing systems, pp 3111–3119Google Scholar
  33. 33.
    Mikolov T, Chen K, Corrado G, Dean J (2013) Efficient estimation of word representations in vector space, computation and languageGoogle Scholar
  34. 34.
    Pang L, Zhu S, Ngo C (2015) Deep multimodal learning for affective analysis and retrieval. IEEE Transactions on Multimedia 17(11):2008–2020CrossRefGoogle Scholar
  35. 35.
    Razzaghnoori M, Sajedi H, Jazani I (2018) Question classification in Persian using word vectors and frequencies. Cogn Syst Res 47:16–27CrossRefGoogle Scholar
  36. 36.
    Rill S, Reinel D, Scheidt J, Zicari R (2014) PoliTwi: early detection of emerging political topics on twitter and the impact on concept-level sentiment analysis. Knowl-Based Syst 69:24–33CrossRefGoogle Scholar
  37. 37.
    Sainath T, Weiss R, Wilson K, Li B, Narayanan A, Variani E, Bacchiani M, Shafran I, Senior A, Chin K, Misra A, Kim C (2017) Multichannel signal processing with deep neural networks for automatic speech recognition. IEEE/ACM Transactions on Audio, Speech, and Language Processing 25(5):965–979CrossRefGoogle Scholar
  38. 38.
    Sangeetha T, Balaganesh N, Muneeswaran K (2017) Aspects based opinion mining from online reviews for product recommendation. In: International conference on computational intelligence in data science(ICCIDS)Google Scholar
  39. 39.
    Saon G, Picheny M (2017) Recent advances in conversational speech recognition using convolutional and recurrent neural networks. IBM J Res Dev 61(4/5):1:1–1:10CrossRefGoogle Scholar
  40. 40.
    Schouten K, Frasincar F (2016) Survey on aspect-level sentiment analysis. IEEE Trans Knowl Data Eng 28(3):813–830CrossRefGoogle Scholar
  41. 41.
    Shams M, Shakery A, Faili H (2012) A non-parametric LDAbased induction method for sentiment analysis. In: The 16th CSI international symposium on artificial intelligence and signal processing (AISP 2012), pp 216–221Google Scholar
  42. 42.
    Socher R, Pennington J, Huang E, Ng A, Manning C (2011) Semi-supervised recursive autoencoders for predicting sentiment distributions. In: Proceedings of the conference on empirical methods in natural language processing. Association for computational linguistics, pp 151–161Google Scholar
  43. 43.
    Socher R, Huval B, Manning CD, Ng AY (2012) Semantic compositionality through recursive matrix-vector spaces. In: Proceeding EMNLP-CoNLL '12 proceedings of the 2012 joint conference on empirical methods in natural language processing and computational natural language learning, pp 1201–1121Google Scholar
  44. 44.
    Sun Y, Wang X, Tang X (2016) Hybrid deep learning for face verification. IEEE Trans Pattern Anal Mach Intell 38(10):1997–2009CrossRefGoogle Scholar
  45. 45.
    Sun S, Luo C, Chen J (2017) A review of natural language processing techniques for opinion mining systems. Information Fusion 36:10–25CrossRefGoogle Scholar
  46. 46.
    Wandabwa H, Asif Naeem M, Mirza F (2017) Document level semantic comprehension of noisy text streams via convolutional neural networks. In: IEEE 21st international conference on computer supported cooperative work in design (CSCWD)Google Scholar
  47. 47.
    Xu X, Cheng X, Tan S, Liu Y, Shen H (2013) Aspect-level opinion mining of online customer reviews. China Communications 10(3):25–41CrossRefGoogle Scholar
  48. 48.
    Zaremba W, Sutskever I (2014) Learning to execute. arXiv preprint arXiv:1410.4615Google Scholar
  49. 49.
    Zhang Y, Wallace B (2015) A sensitivity analysis of (and practitioners’ guide to) convolutional neural networks for sentence classification, computation and language.Google Scholar
  50. 50.
    Zhao R, Mao K (2017) Topic-aware deep compositional models for sentence classification. IEEE/ACM Transactions on Audio, Speech, and Language Processing 25(2):248–260CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Computer Engineering, Faculty of EngineeringShahid Chamran University of AhvazAhvazIran

Personalised recommendations