GenSent: Genetik Algoritma Tabanlı Topluluk Optimizasyonu Kullanılarak Duygu Analizinin İyileştirilmesi

Öz

Sosyal medya platformları, kişisel etkileşimlerden fikir paylaşımına ve önemli uluslararası haberlerin yayılmasına kadar her türlü iletişimin birincil ortamıdır. Bununla birlikte, kullanıcılar tarafından oluşturulan metinsel bilginin sürekli artan miktarı, dilin dinamik yapısı, kullanılan ifadelerdeki ince veya gizli nüanslar ve metindeki bağlamsal bağımlılıklar, zamanında ve doğru duygu analizini giderek daha zorlu hale getirmektedir. Duygu analizi kendi başına önemli bir görevdir ve nefret söylemi ve yanlış bilgi tespiti gibi birçok diğer sınıflandırma görevinin ilk adımı olarak da kullanılır. Duygu analizi ve görüş madenciliği üzerine yapılan araştırmaların önemli bir kısmı, sosyal medya içeriklerini olumlu, olumsuz veya nötr olmak üzere üç sınıfa ayırmaya odaklanmıştır. Ancak, birçok pratik alandaki önemlerine rağmen, son derece olumsuz ve son derece olumlu duygular gibi aşırı görüşlerin sınıflandırılması ancak son zamanlarda ilgi görmeye başlamıştır. Bu boşluğu gidermek için, duygu sınıflandırması için yeni bir genetik algoritma tabanlı optimizasyon çerçevesi olan GenSent adlı bir çerçeve öneriyoruz. Genellikle belirli veri kümelerine göre uyarlanan geleneksel yöntemlerin aksine, GenSent, uç duyguları da temsil eden ikili, üçlü ve ince taneli 5 puanlık ölçek sınıflandırmasından çeşitli duygu analizi görevlerine uygulanabilen çok yönlü bir çerçeve sunar. Destek vektör makineleri, Naïve Bayes, Lojistik Regresyon, Karar Ağaçları, Rastgele Ormanlar ve Stokastik Gradyan İniş Algoritmaları dahil olmak üzere çeşitli sınıflandırıcı havuzlarının kullanımıyla GenSent, herhangi bir müdahale olmaksızın güçlü bir topluluk oluşturur. Çerçeve, ikili, üçlü ve ince taneli duygu analizi veri kümeleri, yani SemEval-2017 (Twitter'da Duygu Analizi) görevi (4A, 4B ve 4C) ve Stanford Duygu Ağaç Bankası (SST-2 ve SST-5) kullanılarak değerlendirilir. Önerilen çerçevenin performansı, aynı veri kümelerini kullanan alanda bilinen diğer mevcut yöntemlerle karşılaştırılır. Karşılaştırmalı sonuçlar, GenSent'in mevcut yöntemlerden daha iyi performans gösterdiğini, hesaplama karmaşıklığını azaltırken çeşitli metriklerde duygu sınıflandırmasında önemli iyileştirmeler sağladığını göstermektedir.

Anahtar Kelimeler

• duygu analizi
• makine öğrenmesi
• optimize edilmiş topluluk sınıflandırıcı
• genetik optimizasyon
• meta-sezgisel algoritmalar.

GenSent: Improving Sentiment Analysis Using Genetic Algorithm-Based Ensemble Optimization

Öz

Social media platforms are currently the primary medium of all types of communication from personal interactions, and opinion sharing to the dissemination of important international news. However, the ever-increasing amount of user-generated textual information coupled with the dynamic nature of the language, subtle or hidden nuances in expressions used, and contextual dependencies in text, renders timely and accurate sentiment analysis increasingly challenging. Sentiment analysis is an important task in its own right and is also used as the first step of many other classification tasks such as hate speech and misinformation detection. A significant portion of research on sentiment analysis and opinion mining has concentrated on categorizing social media content into three classifications: positive, negative, or neutral. However, despite their importance across numerous practical domains, the classification of extreme opinions, such as highly negative and highly positive sentiments, has only recently gained attention. To address this gap, we propose a framework, GenSent, a novel genetic algorithm-based optimization framework for sentiment classification. Unlike traditional methods that are often tailored to specific datasets, GenSent provides a versatile framework applicable to diverse sentiment analysis tasks from binary, ternary, and fine-grained 5-point scale classification that represents extreme sentiments as well. Through the use of a diverse pool of classifiers including support vector machines, Naïve Bayes, Logistic Regression, Decision Trees, Random Forests, and Stochastic Gradient Descent Algorithms, GenSent effectively builds a robust ensemble without any intervention. The framework is evaluated using binary, ternary, and fine-grained sentiment analysis datasets, namely, SemEval-2017 (Sentiment Analysis in Twitter) task (4A, 4B, and 4C) and Stanford Sentiment Treebank (SST-2 and SST-5). The performance of the proposed framework is compared with other existing well-known methods in the field using the same datasets. Comparative results demonstrate that GenSent outperforms existing methods, achieving significant improvements in sentiment classification across various metrics while reducing the computational complexity.

Anahtar Kelimeler

• sentiment analysis
• machine learning
• optimized ensemble classifier
• genetic optimization
• meta-heuristic algorithms.

Kaynakça

1. [1] Alarifi, A., Alsaleh, M., and Al-Salman, A., “Twitter turing test: Identifying social machines”, Information Sciences, 372: 332-346, (2016).
2. [2] Öztürk, N., and Ayvaz, S., “Sentiment analysis on Twitter: A text mining approach to the Syrian refugee crisis.”, Telematics and Informatics, 35(1): 136-147, (2018).
3. [3] Liu, B., “Sentiment analysis and opinion mining.”, Synthesis lectures on human language technologies, 5(1): 1-167, (2012).
4. [4] Kour, H., and Gupta, M. K., “Hybrid evolutionary intelligent network for sentiment analysis using Twitter data during COVID‐19 pandemic.”, Expert Systems, 41(3): e13489, (2024).
5. [5] Bird, S., Klein, E., and Loper, E., "Natural language processing with Python: analyzing text with the natural language toolkit.”, O’Reilly Media, Inc., (2009).
6. [6] Pang, B., Lee, L., and Vaithyanathan, S., “Thumbs up?: sentiment classification using machine learning techniques.”, In Proceedings of the ACL-02 conference on Empirical methods in natural language processing- Association for Computational Linguistics, 10: 79-86, (2002).
7. [7] Turney, P. D., “Thumbs up or thumbs down?: semantic orientation applied to unsupervised classification of reviews.”, In Proceedings of the 40th annual meeting on association for computational linguistics- Association for Computational Linguistics, 417-424, (2002).
8. [8] Stoyanov, V., & Cardie, C., “Topic identification for fine-grained opinion analysis.”, In Proceedings of the 22nd International Conference on Computational Linguistics, Coling, 817-824, (2008).

9. [9] Kouloumpis, E., Wilson, T., and Moore, J., “Twitter sentiment analysis: The good the bad and the omg!.”, In Fifth International AAAI conference on weblogs and social media, Edinburgh, (2011).
10. [10] Villena-Román, J., Lana-Serrano, S., Martínez-Cámara, E., and González-Cristóbal, J. C., “Tass-workshop on sentiment analysis at sepln.”, Procesamiento del Lenguaje Natural, 50: 37-44, (2013).
11. [11] Go, A., Bhayani, R., & Huang, L., “Twitter sentiment classification using distant supervision.”, CS224N project report, Stanford, 1:12, (2009).
12. [12] Nakov, P., Rosenthal, S., Kozareva, Z., Stoyanov, V., Ritter, A., and Wilson, T., “SemEval-2013 Task 2: Sentiment Analysis in Twitter.”, In Second Joint Conference on Lexical and Computational Semantics (* SEM), Volume 2: Proceedings of the Seventh International Workshop on Semantic Evaluation (SemEval 2013), 312-320, (2013).
13. [13] Rosenthal , S., Ritter , A., Nakov , P., and Stoyanov , V., “SemEval-2014 Task 9: Sentiment Analysis in Twitter.”, In Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), 73–80, Dublin, Ireland, (2014).
14. [14] Nakov, P., Rosenthal, S., Kiritchenko, S., Mohammad, S. M., Kozareva, Z., Ritter, A., ... & Zhu, X., “Developing a successful SemEval task in sentiment analysis of Twitter and other social media texts.”, Language Resources and Evaluation, 50: 35-65, (2016).
15. [15] Rosenthal, S., Nakov, P., Kiritchenko, S., Mohammad, S., Ritter, A., Stoyanov, V., “SemEval-2015 task 10: Sentiment analysis in Twitter.”, In: Proceedings of the 9th International Workshop on Semantic Evaluation, SemEval '15, 450–462, Denver, Colorado, USA, (2015).
16. [16] Nakov, P., Ritter, A., Rosenthal, S., Sebastiani, F., and Stoyanov, V., “SemEval-2016 task 4: Sentiment analysis in Twitter.”, arXiv preprint arXiv:1912.01973, (2019).
17. [17] Rosenthal, S., Farra, N., and Nakov, P., “SemEval-2017 task 4: Sentiment analysis in Twitter.”, In Proceedings of the 11th international workshop on semantic evaluation (SemEval-2017), 502-518, (2017).
18. [18] Gamal, D., Alfonse, M., M El-Horbaty, E. S., & M Salem, A. B., “Analysis of Machine Learning Algorithms for Opinion Mining in Different Domains.”, Machine Learning and Knowledge Extraction, 1: 224-234, (2019).
19. [19] Liu, X. Y., Zhang, K. Q., Fiumara, G., Meo, P. D., and Ficara, A., “Adaptive Evolutionary Computing Ensemble Learning Model for Sentiment Analysis.”, Applied Sciences, 14: 6802, (2024).
20. [20] Sar-Saifee, B., Tanha, J., & Aeini, M., “A Hybrid Deep Learning Network for Sentiment Analysis on SemEval-2017 Dataset.”, In 2023 28th International Computer Conference, Computer Society of Iran (CSICC), 1-7, IEEE, (2023).
21. [21] Arif, F., and Dulhare, U. N., “A Machine Learning Based Approach for Opinion Mining on Social Network Data.”, In Computer Communication, Networking and Internet Security, 135-147, Springer, Singapore, (2017).
22. [22] Gogna, A., and Tayal, A., “Metaheuristics: review and application.”, Journal of Experimental & Theoretical Artificial Intelligence, 25: 503-526, (2013).
23. [23] Dos Santos, E. M., “Evolutionary algorithms applied to classifier ensemble selection.”, XLIV SBPO/XVI CLAIO, 419-430, (2012).
24. [24] Symeonidis, S., Effrosynidis, D., Kordonis, J., & Arampatzis, A., “DUTH at SemEval-2017 Task 4: a voting classification approach for Twitter sentiment analysis.”, In Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017), 704-708, (2017).
25. [25] Hasan, A., Moin, S., Karim, A., and Shamshirband, S., “Machine learning-based sentiment analysis for twitter accounts.”, Mathematical and Computational Applications, 23: 11, (2018).
26. [26] Othman, M., Hassan, H., Moawad, R., and Idrees, A. M., “A linguistic approach for opinionated documents summary.”, Future Computing and Informatics Journal, 3:152-158, (2018).
27. [27] Wang, J., and Dong, A., “A comparison of two text representations for sentiment analysis.”, In 2010 International Conference on Computer Application and System Modeling (ICCASM 2010), 11, V11-35, IEEE, (2010).
28. [28] Kanayama, H., Nasukawa, T., and Watanabe, H., “Deeper sentiment analysis using machine translation technology.”, In COLING 2004: Proceedings of the 20th International Conference on Computational Linguistics, 494-500, (2004).
29. [29] Raychev, V., and Nakov, P., “Language-independent sentiment analysis using subjectivity and positional information.”, arXiv preprint arXiv:1911.12544, (2019).
30. [30] Das, R. K., & Pedersen, D. T., “SemEval-2017 Task 4: Sentiment Analysis in Twitter using BERT.”, arXiv preprint arXiv:2401.07944, (2024).
31. [31] Younesi, R. T., Tanha, J., Namvar, S., and Mostafaei, S. H., “A CNN-BiLSTM based deep learning model to sentiment analysis.”, In 2024 20th CSI International Symposium on Artificial Intelligence and Signal Processing (AISP), 1-6, IEEE, (2024).
32. [32] Dimililer, N., Varoğlu, E., and Altınçay, H., “Vote-based classifier selection for biomedical NER using genetic algorithms.”, In Iberian Conference on Pattern Recognition and Image Analysis, 202-209, Springer, Berlin, Heidelberg, (2007).
33. [33] Dimililer, N., Varoğlu, E., and Altınçay, H., “Classifier subset selection for biomedical named entity recognition.”, Applied Intelligence, 31(3):267-282, (2009).
34. [34] Park, D., and Ahn, C. W., “Self-Supervised Contextual Data Augmentation for Natural Language Processing.”, Symmetry, 11(11): 1393, (2019).
35. [35] Xu, Y., Li, L., Gao, H., Hei, L., Li, R., and Wang, Y., “Sentiment classification with adversarial learning and attention mechanism.”, Computational Intelligence, 37(2): 774-798, (2021).
36. [36] Kasri, M., Birjali, M., and Beni‐Hssane, A., “Word2Sent: A new learning sentiment‐embedding model with low dimension for sentence level sentiment classification.”, Concurrency and Computation: Practice and Experience, 33(9): e6149, (2021).
37. [37] Sadr, H., Solimandarabi, M. N., Pedram, M. M., and Teshnehlab, M., “A Novel Deep Learning Method for Textual Sentiment Analysis.”, arXiv preprint arXiv:2102.11651, (2021).
38. [38] Giménez, M., Palanca, J., and Botti, V., “Semantic-based padding in convolutional neural networks for improving the performance in natural language processing. A case of study in sentiment analysis.”, Neurocomputing, 378: 315-323, (2020).
39. [39] Tripathi, S., Singh, C., Kumar, A., Pandey, C., and Jain, N., “Bidirectional transformer based multi-task learning for natural language understanding.”, In International Conference on Applications of Natural Language to Information Systems, 54-65, Springer, Cham, (2019).
40. [40] Hama Aziz, R. H., and Dimililer, N., “SentiXGboost: enhanced sentiment analysis in social media posts with ensemble XGBoost classifier.”, Journal of the Chinese Institute of Engineers, 1-11, (2021).
41. [41] Kuncheva, L. I., “Combining pattern classifiers: methods and algorithms.”, John Wiley & Sons, (2014).
42. [42] Kirasich, K., Smith, T., and Sadler, B., “Random Forest vs logistic regression: binary classification for heterogeneous datasets.”, SMU Data Science Review, 1(3): 9, (2018).
43. [43] Breiman, L., “Random forests.”, Machine learning, 45(1): 5-32, (2001).
44. [44] Hosmer Jr, D. W., Lemeshow, S., and Sturdivant, R. X., “Applied logistic regression”, John Wiley & Sons, 398, (2013).
45. [45] Al Amrani, Y., Lazaar, M., and El Kadiri, K. E., “Random forest and support vector machine based hybrid approach to sentiment analysis.”, Procedia Computer Science, 127: 511-520, (2018).
46. [46] Kraft, D. H., Petry, F. E., Buckles, B. P., and Sadasivan, T., “The use of genetic programming to build queries for information retrieval.”, In Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence, 468-473, IEEE, (1994).
47. [47] Martin-Bautista, M. J., Larsen, H. L., Nicolaisen, J., and Svendsen, T., “An approach to an adaptive information retrieval agent using genetic algorithms with fuzzy set genes.”, In Proceedings of 6th International Fuzzy Systems Conference, 3: 1227-1232, IEEE, (1997).
48. [48] Ishaq, A., Asghar, S., and Gillani, S. A., “Aspect-Based Sentiment Analysis Using a Hybridized Approach Based on CNN and GA.”, IEEE Access, 8: 135499-135512, (2020).
49. [49] Cahya, R. A., Adimanggala, D., and Supianto, A. A., “Deep Feature Weighting Based on Genetic Algorithm and Naïve Bayes for Twitter Sentiment Analysis.”, In 2019 International Conference on Sustainable Information Engineering and Technology (SIET), 326-331, IEEE, (2019).
50. [50] Iqbal, F., Hashmi, J. M., Fung, B. C., Batool, R., Khattak, A. M., Aleem, S., and Hung, P. C., “A hybrid framework for sentiment analysis using genetic algorithm based feature reduction.”, IEEE Access, 7: 14637-14652, (2019).
51. [51] Fatyanosa, T. N., Bachtiar, F. A., and Data, M., “Feature Selection using Variable Length Chromosome Genetic Algorithm for Sentiment Analysis.”, In 2018 International Conference on Sustainable Information Engineering and Technology (SIET), 27-32, IEEE, (2018).
52. [52] Keshavarz, H., Abadeh, M. S., and Almasi, M., “A new lexicon learning algorithm for sentiment analysis of big data.”, In 2017 IEEE 15th International Symposium on Intelligent Systems and Informatics (SISY), 000249-000254, IEEE, (2017).
53. [53] Saidani, F. R., and Rassoul, I., “A weighted genetic approach for feature selection in sentiment analysis.”, International Journal of Computational Intelligence and Applications, 16(02): 1750013, (2017).
54. [54] Baziotis, C., Pelekis, N., and Doulkeridis, C., “Datastories at semeval-2017 task 4: Deep lstm with attention for message-level and topic-based sentiment analysis.”, In Proceedings of the 11th international workshop on semantic evaluation (SemEval-2017), 747-754, (2017).
55. [55] Cliche, M., “Bb_twtr at semeval-2017 task 4: Twitter sentiment analysis with cnns and lstms.”, arXiv preprint, arXiv:1704.06125, (2017).
56. [56] Kolovou, A., Kokkinos, F., Fergadis, A., Papalampidi, P., Iosif, E., Malandrakis, N., Palogiannidi, E., Papageorgiou, H., Narayanan, S. and Potamianos, A., “Tweester at SemEval-2017 Task 4: Fusion of Semantic-Affective and pairwise classification models for sentiment analysis in Twitter.”, In Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017), 675-682, (2017).
57. [57] Yadav, N., Kudale, O., Gupta, S., Rao, A., and Shitole, A., “Twitter Sentiment Analysis Using Machine Learning for Product Evaluation.”, In 2020 International Conference on Inventive Computation Technologies (ICICT), 181-185, IEEE, (2020).
58. [58] Moh, M., Gajjala, A., Gangireddy, S. C. R., and Moh, T. S., “On multi-tier sentiment analysis using supervised machine learning.”, In 2015 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT), 1: 341-344, IEEE, (2015).
59. [59] Dong, Y., Fu, Y., Wang, L., Chen, Y., Dong, Y., and Li, J., “A sentiment analysis method of capsule network based on BiLSTM.”, IEEE Access, 8: 37014-37020, (2020).
60. [60] Hassan, A., and Mahmood, A., “Deep learning approach for sentiment analysis of short texts.”, In 2017 3rd international conference on control, automation and robotics (ICCAR), 705-710, IEEE, (2017).
61. [61] Li, W., Zhu, L., Shi, Y., Guo, K., and Zheng, Y., “User reviews: Sentiment analysis using lexicon integrated two-channel CNN-LSTM family models.”, Applied Soft Computing, 106435, (2020).
62. [62] Hiyama, Y., and Yanagimoto, H., “Word polarity attention in sentiment analysis.”, Artificial Life and Robotics, 23(3): 311-315, (2018).
63. [63] Baktha, K., and Tripathy, B. K., “Investigation of recurrent neural networks in the field of sentiment analysis.”. In 2017 International Conference on Communication and Signal Processing (ICCSP), 2047-2050, IEEE, (2017).
64. [64] Chen, T., Xu, R., He, Y., and Wang, X., “Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN.”, Expert Systems with Applications, 72: 221-230, (2017).
65. [65] Sadr, H., Pedram, M. M., and Teshnehlab, M., “A robust sentiment analysis method based on sequential combination of convolutional and recursive neural networks.”, Neural Processing Letters, 50(3): 2745-2761, (2019).
66. [66] Lu, Y., Rao, Y., Yang, J., and Yin, J., “Incorporating Lexicons into LSTM for sentiment classification.’, In 2018 International joint conference on neural networks (IJCNN), 1-7, IEEE, (2018).
67. [67] Lei, Z., Yang, Y., and Yang, M. , SAAN: A sentiment-aware attention network for sentiment analysis. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval 1197-1200. (2018).
68. [68] George, C. S., and Sumathi, B., “Genetic Algorithm Based Hybrid Model of Convolutional Neural Network And Random Forest Classifier For Sentiment Classification.”, Turkish Journal of Computer and Mathematics Education, 12(2): 3216-3223, (2021).
69. [69] Huang, J. Y., Tung, C. L., and Lin, W. Z., “Using social network sentiment analysis and genetic algorithm to improve the stock prediction accuracy of the deep learning-based approach.”, International Journal of Computational Intelligence Systems, 16(1): 93, (2023).
70. [70] Jain, A., and Jain, V., “Sentiment classification using hybrid feature selection and ensemble classifier.”, Journal of Intelligent & Fuzzy Systems, 42(2): 659-668, (2022).
71. [71] Nkhata, G., & Gauch, S.," Fine-tuning BERT with Bidirectional LSTM for Find-Grained Movie Reviews Sentiment Analysis". International Journal On Advances in Systems and Measurements. (2023)
72. [72] Wang, J., Zhang, Y., Yu, L. C., and Zhang, X., “Contextual sentiment embeddings via bi-directional GRU language model.”, Knowledge-Based Systems, 235: 107663, (2022).
73. [73] Cao, B., Jiang, K., and Fan, J., “SLaNT: A Semi-supervised Label Noise-Tolerant Framework for Text Sentiment Analysis.”, In Proceedings of the International AAAI Conference on Web and Social Media, 18: 191-202, (2024).
74. [74] Aydın, N., Erdem, O. A., and Tekerek, A., “Comparative analysis of traditional machine learning and transformer-based deep learning models for text classification.”, Journal of Polytechnic (Politeknik Dergisi), 28(2): 445-452, (2025).
75. [75] Tunç, Ü., Atalar, E., Gargı, M. S., and Aydın, Z. E., “Classification of fake, bot, and real accounts on instagram using machine learning.”, Politeknik Dergisi, 27(2): 479-488, (2022).
76. [76] [76] Hassanat, A., Almohammadi, K., Alkafaween, E. A., Abunawas, E., Hammouri, A., and Prasath, V. S., “Choosing mutation and crossover ratios for genetic algorithms—a review with a new dynamic approach.”, Information, 10(12): 390, (2019).
77. [77] Azedou, A., Amine, A., Kisekka, I., and Lahssini, S., “Genetic algorithm optimization of ensemble learning approach for improved land cover and land use mapping: Application to Talassemtane National Park.”, Ecological Indicators, 177: 113776, (2025).
78. [78] Huang, J. Y., Tung, C. L., and Lin, W. Z., “Using social network sentiment analysis and genetic algorithm to improve the stock prediction accuracy of the deep learning-based approach.”, International Journal of Computational Intelligence Systems, 16(1): 93, (2023).
79. [79] Demirel, U., and Çam, H., “Investigation of fluctuations incryptocurrency transactions with sentiment analysis.” Politeknik Dergisi, 28(3): 773–784, (2025).
80. [80] Aziz, R. H. H., and Dimililer, N., “Twitter sentiment analysis using an ensemble weighted majority vote classifier.”, In 2020 International Conference on Advanced Science and Engineering (ICOASE), 103-109, IEEE, (2020).