یک مدل عمیق CNN-BiLSTM مبتنی بر مکانیزم توجه برای تحلیل احساسات نظرات کاربران در رسانه‌های اجتماعی

نوع مقاله : علمی-پژوهشی

نویسندگان

1 گروه مهندسی کامپیوتر- دانشکده مهندسی برق و کامپیوتر- دانشگاه تبریز- تبریز- ایران

2 دانشکده مهندسی برق و کامپیوتر- دانشگاه تبریز- تبریز - ایران

3 گروه مهندسی مخابرات- دانشکده مهندسی برق و کامپیوتر- دانشگاه تبریز- تبریز - ایران

چکیده

هدف از تحلیل احساسات، تعیین جهتی برای بیان دیدگاه و نظرات کاربران و مشاهده وضعیت عمومی با توجه به هوش اقتصادی، سنجش رضایت مشتری، پیش بینی فروش فیلم و موارد دیگر می باشد. همچنین تجزیه و تحلیل احساسات، به عنوان یکی از زمینه‌های پردازش زبان طبیعی شناخته می‌شود. هدف از این پژوهش، ارائه رویکردی جدید برای تجزیه و تحلیل احساسات و تشخیص قطبیت نظرات کاربران رسانه‌های اجتماعی با استفاده از الگوریتم‌های یادگیری عمیق است.
در این مطالعه با استفاده از تلفیق شبکه‌های CNN و Bi-LSTM و با به کارگیری مکانیزم توجه مدلی پیشنهاد می‌شودکه از CNN برای استخراج ویژگی متن و از شبکهBi-LSTM برای طبقه‌بندی احساسات استفاده می‌کند. مکانیزم توجه نیز می‌تواند بر بخش‌های مهم یک دنباله تمرکز کرده و در نتیجه کارایی شبکه‌های عصبی را در کارهای مختلف از جمله تجزیه و تحلیل احساسات، ترجمه ماشینی و تشخیص گفتار افزایش ‌دهد. نتایج بدست آمده نشان می‌دهد که روش پیشنهادی، دقت بالاتر و پیچیدگی کمتری برای تعیین قطبیت نظرات کاربران در رسانه‌های اجتماعی نسبت به روش‌های موجود در تحقیقات اخیر دارد.

کلیدواژه‌ها

عنوان مقاله [English]

Attention-based CNN-BiLSTM Deep Model for Sentiment Analysis of User Opinions in Social Media

نویسندگان [English]

  • Reza Latifi Ardahaei 1
  • Mohammad Asadpour 2
  • Hadi Seyedarabi 3
1 Computer Department, Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran
2 Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran
3 Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran
چکیده [English]

Extracting sentiment from textual data is crucial for understanding public opinion and guiding strategic decisions. We introduce a hybrid deep-learning pipeline that combines convolutional feature detectors, bidirectional recurrent units, and a custom attention mechanism. First, convolutional layers with pooling condense local n-gram patterns into compact feature maps. These maps are fed into a bidirectional LSTM network that captures sequence information in both forward and reverse directions. A specialized attention module then assigns relevance scores to individual tokens, sharpening the final classification. Evaluations on widely used sentiment benchmarks show that our method outperforms leading models in terms of accuracy as well as requires fewer computational resources, making it a practical solution for scalable emotion detection in text.

کلیدواژه‌ها [English]

  • Sentiment analysis
  • Deep learning
  • CNN
  • Bi-LSTM
  • Attention mechanism
  • Social media

مراجع

[1] A. Kumar Sharma, S. Chaurasia and D. Kumar Srivastava, "Sentimental Short Sentences Classification by Using CNN Deep Learning Model with Fine Tuned Word2Vec," Procedia Computer Science, vol. 167, pp. 1139-1147, April 2020.
[2] J. S. and . M. M. , "An efficient sentiment analysis methodology based on long short-term memory networks," Complex & Intelligent Systems, vol. 7, p. 2485–2501, 18 June 2021.
[3] N. Cach Dang , M. N. Moreno-García and F. De la Prieta, "Sentiment Analysis Based on Deep Learning:A Comparative Study," Electronics, 14 March 2020.
[4] M. Birjali, M. Kasri and A. Beni-Hssane, "A comprehensive survey on sentiment analysis: Approaches,challenges and trends," Knowledge-Based Systems, vol. 226, p. 107134, 17 August 2021.
[5]  M. D. and . R. R. , "An efficient sentimental analysis using hybrid deep learning and optimization technique for Twitter using parts of speech (POS) tagging," International Journal of Speech Technology, vol. 24, p. 329–339, 1 February 2021.
[6] Zafarani-Moattar, E., Kangavari, M. R., & Rahmani, A. M. (2022). Topic detection on COVID-19 tweets: A comparative study on clustering and transfer learning models. Tabriz Journal of Electrical Engineering, 52(4), 281–291. https://doi.org/10.22034/tjee.2022.15744.
[7] R. Kumar Behera, M. Jena, S. Kumar Rath and S. Misra, "Co-LSTM: Convolutional LSTM model for sentiment analysis in social big data," Information Processing and Management, vol. 58, no. 1, January 2021.
[8] S. Sachin, A. Tripathi, N. Mahajan, S. Aggarwal and P. Nagrath, "Sentiment Analysis Using Gated Recurrent Neural Networks," SN Computer Science, March 2020.
[9] S. TAM, R. BEN SAID and Ö. ÖZGÜR TANRIÖVER, "A ConvBiLSTM Deep Learning Model-Based Approach for Twitter Sentiment Classification," IEEE Access, vol. 9, pp. 41283 - 41293, 9 March 2021.
[10] S. Kardakis, I. Perikos, F. Grivokostopoulou and I. Hatzilygeroudis, "Examining Attention Mechanisms in Deep Learning Models for Sentiment Analysis," Applied Sciences 11, no. 9, 25 April 2021.
[11] Z. Yang, Z. Dai, Y. Yang1, J. Carbonell1, R. Salakhutdinov1 and Q. V. Le, "XLNet: Generalized Autoregressive Pretraining for Language Understanding," January 2020.
[12] M. Umut Salur and I. Aydin, "A Novel Hybrid Deep Learning Model for Sentiment Classification," IEEE Access, vol. 8, pp. 58080 - 58093, 23 March 2020.
[13] X. FU, J. YANG, J. LI, M. FANG3 and H. WANG4, "Lexicon-enhanced LSTM with Attention for General Sentiment Analysis," IEEE Access, vol. 6, pp. 71884 - 71891, 29 October 2018.
[14] M. E. Basiri, S. Nemati, M. Abdar and E. Cambria, "ABCDM: An Attention-based Bidirectional CNN-RNN Deep Model for sentiment analysis," Future Generation Computer Systems, vol. 115, pp. 279-294, February 2021.
[15] C. N. Dang, M. N. Moreno-Garcı´a and F. De la Prieta, "Hybrid Deep Learning Models for Sentiment Analysis," Complexity, 13 August 2021.
[16] G. XU, Y. MENG, X. QIU, Z. YU and X. WU, "Sentiment Analysis of Comment Texts Based on BiLSTM," IEEE Access, vol. 7, pp. 51522 - 51532, 9 April 2019.
[17] A. H. Ombabi, O. Lazzez, W. Ouarda and A. M. Alimi, "Deep learning framework based on Word2Vec and CNNfor users interests classification," in Sudan Conference on Computer Science and Information Technology (SCCSIT), 2017.
 [18] پورحسینی، محمدرضا؛ عباسی، مهدی و محمدی‌پسند، احسان. (۱۴۰۳). توسعه مدل‌های کوچک یادگیری ماشین برای توزیع بهینه بارهای کاری در شبکه‌های لبه. مجله مهندسی برق دانشگاه تبریز، ۵۴(۴)،
 ۴۰۴-۴۱۳ https://doi.org/10.22034/tjee.2024.57664.4672.
[19] G. Liu and J. Guo, "Bidirectional LSTM with attention mechanism and convolutional layer for text classification," Neurocomputing, vol. 337, pp. 325-338, 14 April 2019.
[20] Z. JIANQIANG, G. XIAOLIN and Z. XUEJUN, "Deep Convolution Neural Networks for Twitter Sentiment Analysis," IEEE Access, vol. 6, pp. 23253 - 23260, 1 January 2018.
[21] Y. LIN, J. LI, L. YANG, K. XU and H. LIN, "Sentiment Analysis With Comparison Enhanced Deep Neural Network," IEEE Access, vol. 8, pp. 78378 - 78384, 22 April 2020.
[22] M. Kamyab, G. Liu and M. Adjeisah, "Attention-Based CNN and Bi-LSTM Model Based on TF-IDF and GloVe Word Embedding for Sentiment Analysis," Applied Sciences, no. 23, 27 November 2021.
[23] H. Tien Nguyen and M. Le Nguyen, "An ensemble method with sentiment features and clustering support," Neurocomputing, vol. 370, pp. 155-165, 22 December 2019.
[24] M. Kamkarhaghighi and M. Makrehchi, "Content Tree Word Embedding for document representation," Expert Systems with Applications, pp. 241-249, 30 December 2017.
[25] S. Wen and J. Li, "Recurrent Convolutional Neural Network with Attention for Twitter and Yelp Sentiment Classification," in Proceedings of the 2018 International Conference on Algorithms, Computing and Artificial Intelligence, 2018.
[26] D. ZHANG, L. TIAN, M. HONG, F. HAN, Y. REN and Y. CHEN, "Combining Convolution Neural Network and Bidirectional Gated Recurrent Unit for Sentence Semantic Classification," IEEE Access, vol. 6, pp. 73750 - 73759, 22 November 2018.
[27] U. Mohd, B. Ahmad, E. Song, M. S. Hossain, M. Alrashoud and G. Muhammad, "Attention-based sentiment analysis using convolutional and recurrent neural network," Future Generation Computer Systems, pp. 571-578, December 2020.
[28] W. Li, F. Qi, M. Tang and Z. Yu, "Bidirectional LSTM with self-attention mechanism and multi-channel features for sentiment classification," Neurocomputing, vol. 387, pp. 63-77, 28 April 2020.
[29] G. Rao, W. Huang, Z. Feng and Q. Cong, "LSTM with sentence representations for document-level sentiment classification," Neurocomputing, vol. 308, pp. 49-57, 25 September 2018.
[30] J. DU, L. GUI, Y. HE, R. XU and X. WANG, "Convolution-Based Neural Attention With Applications to Sentiment Classification," IEEE Access, vol. 7, pp. 27983 - 27992, 21 February 2019.
[31] A. U. Rehman, A. K. Malik, B. Raza and W. Ali, "A Hybrid CNN-LSTM Model for Improving Accuracy of Movie Reviews Sentiment Analysis," Multimedia Tools and Applications, vol. 78, p. 26597–26613, 11 June 2019.
[32] B. Ahmed Chandio, A. Shariq Imran, M. Bakhtyar, S. M. Daudpota and J. Baber, "Attention-Based RU-BiLSTM Sentiment Analysis Model for Roman Urdu," Applied Sciences, no. 7, 4 April 2022.
[33] M. Mishra, A. Patil, " Sentiment Prediction of IMDb Movie Reviews Using CNN-LSTM Approach", International Conference on Control Communication & Computing India (ICCC) , July 2023.
[34] S. A. Sazan, M. Ahmed, T. B. Saad, M. Roy "Advanced Natural Language Processing Techniques for Efficient Sentiment Analysis of US Airline Twitter Data: A High-Performance Framework for Extracting Insights from Tweets", 2024 6th International Conference on Electrical Engineering and Information & Communication Technology (ICEEICT).
[35] R. Marino, L. Buffoni, L. Chicchi, F. D. Patti, D. Febbe, L. Giambagli, D. Fanelli, " Learning in Wilson-Cowan model for metapopulation", 2024.

فایل ها

هم رسانی

ارجاع به این مقاله

آمار