Türkçe metinlerde hedef terimi çıkarımı için bir topluluk yaklaşımı

Öz

Günümüzde standart duygu analizinin yetersiz kalması sonucunda, hedef tabanlı duygu analizi (HTDA) çalışmaları büyük ilgi görmüştür. HTDA, bir metindeki her terim/nitelik hakkında ayrıntılı duygu ve düşüncelerin ortaya çıkarılmasını sağlar. HTDA yönteminin en önemli alt aşaması, bir metinden hedef terimlerinin çıkarılması işlemidir. Türkçe gibi sondan eklemeli dil yapılarına sahip metinlerde bu süreç daha da zorlaşmaktadır. Bu çalışmada, Türkçe kullanıcı yorumlarından hedef terimlerini çıkarmak için istatistiksel (TF-IDF), konu modelleme (LDA ve NMF) ve kural-tabanlı yöntemleri bir arada kullanan bir topluluk yaklaşımı önerilmiştir. Önerilen yöntem, farklı yöntemlerle elde edilen aday hedef terim kümelerini stratejik olarak birleştirir ve nihai hedef terimleri listesini belirler. Önerilen yöntem, Türkçe restoran yorumlarından oluşan SemEval-2016 HTDA kıyaslama veri seti üzerinde test edilmiştir. Önerilen yöntemin deneysel sonuçları aynı veri kümesi üzerinde yapılan önceki çalışmalarla karşılaştırılmıştır.

Anahtar Kelimeler

• Hedef terim çıkarımı
• Hedef tabanlı duygu analizi
• Türkçe metinler
• Duygu analizi

An ensemble approach for aspect term extraction in Turkish texts

Öz

Today, as a result of the inadequacies of the standard sentiment analysis, aspect-based sentiment analysis (ABSA) studies have great attracting interest. ABSA reveals detailed sentiment and opinion about every term/attribute in a text. The most important sub-stage of the ABSA method is the process of extracting the aspect terms from a text. This process becomes more difficult in texts with agglutinative language structures such as Turkish. In this study, we proposed an ensemble approach that uses statistical (TF-IDF), topic modeling (LDA and NMF), and rule-based methods together to extract aspect terms from Turkish user comments. The proposed method strategically combines the candidate aspect term obtained by different methods and determines the final aspect term lists. The proposed method has been tested on the SemEval-2016 ABSA benchmarking dataset, which consists of Turkish restaurant reviews. The experimental results of the proposed method were compared with previous studies on the same dataset.

Anahtar Kelimeler

• Aspect term extraction
• Aspect based sentiment analysis
• Turkish texts
• Sentiment analysis

Kaynakça

1. [1] Xianghua F, Guo L, Yanyan G, Zhiqiang W. “Multi-aspect sentiment analysis for Chinese online social reviews based on topic modeling and HowNet lexicon”. Knowledge-Based Systems, 37, 186-195, 2013.
2. [2] Augustyniak Ł, Kajdanowicz T, Kazienko P. “Comprehensive analysis of aspect term extraction methods using various text embeddings”. Computer Speech & Language, 69, 1-19, 2021.
3. [3] Salur MU, Aydin I. “A novel hybrid deep learning model for sentiment classification”. IEEE Access, 8, 58080-58093, 2020.
4. [4] Çoban Ö, Özyer GT. “Twitter duygu analizinde terim ağırlıklandırma yönteminin etkisi”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 24(2), 283-291, 2018.
5. [5] Pontiki M, Galanis D, Pavlopoulos J, Papageorgiou H, Androutsopoulos I, Manandhar S. “SemEval-2014 Task 4: aspect based sentiment analysis”. 8th International Workshop on Semantic Evaluation (SemEval 2014), Dublin, Ireland, 8 August 2014.
6. [6] Pontiki M, Galanis D, Papageorgiou H, Manandhar S, Androutsopoulos I. “SemEval-2015 Task 12: Aspect Based Sentiment Analysis”. 9th international workshop on semantic evaluation (SemEval 2015), Denver, Colorado, 4-5 June 2015.
7. [7] Ansari G, Saxena C, Ahmad T, Doja MN. “Aspect Term Extraction using graph-based semi-supervised learning”. Procedia Computer Science, 167, 2080-2090, 2020.
8. [8] Wang W, Pan SJ, Dahlmeier D, Xiao X. “Recursive neural conditional random fields for aspect-based sentiment analysis”. arXiv, 2016. https://arxiv.org/pdf/1603.06679.pdf

9. [9] Bhamare BR, Prabhu J. “A supervised scheme for aspect extraction in sentiment analysis using the hybrid feature set of word dependency relations and lemmas”. PeerJ Computer Science, 7, 1-22, 2021.
10. [10] Poria S, Cambria E, Gelbukh A. “Aspect extraction for opinion mining with a deep convolutional neural network”. Knowledge-Based Systems, 108, 42-49, 2016.
11. [11] Luo H, Li T, Liu B, Wang B, Unger H. “Improving aspect term extraction with bidirectional dependency tree representation.” IEEE/ACM Transactions on Audio, Speech, and Language Processing, 27(7), 1201-1212, 2019.
12. [12] Perikos I, Hatzilygeroudis I. “Aspect based sentiment analysis in social media with classifier ensembles”. 2017 IEEE/ACIS 16th International Conference on Computer and Information Science (ICIS), Wuhan, China, 24-26 May 2017.
13. [13] Manek AS, Shenoy PD, Mohan MC, Venugopal KR. “Aspect term extraction for sentiment analysis in large movie reviews using Gini Index feature selection method and SVM classifier”. World Wide Web, 20(2), 135-154, 2017.
14. [14] Wen H, Zhao J. “Aspect term extraction of E-commerce comments based on model ensemble”. 14th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP), Chengdu, China, 15-17 December 2017.
15. [15] Akhtar MS, Gupta D, Ekbal A, Bhattacharyya P. “Feature selection and ensemble construction: A two-step method for aspect based sentiment analysis”. Knowledge-Based Systems, 125, 116-135, 2017.
16. [16] Chauhan GS, Kumar YM.“Prominent Aspect Term Extraction in Aspect Based Sentiment Analysis”. 3rd International Conference and Workshops on Recent Advances and Innovations in Engineering (ICRAIE), Jaipur, India, 22-25 November 2018.
17. [17] Luo Z, Huang S, Zhu KQ. “Knowledge empowered prominent aspect extraction from product reviews”. Information Processing & Management, 56(3), 408-423, 2019.
18. [18] Chauhan GS, Meena YK, Gopalani D, Nahta R. “A two-step hybrid unsupervised model with attention mechanism for aspect extraction”. Expert Systems with Applications, 161, 1-14, 2020.
19. [19] Dehkharghani R, Yanikoglu BA, Saygin Y, Oflazer K. “Sentiment analysis in Turkish at different granularity levels”. Natural Language Engineering, 23(4), 535-559, 2017.
20. [20] Türkmen H, Omurca SI, Ekinci E. “An aspect based sentiment analysis on Turkish hotel reviews”. Girne American University Journal of Social and Applied Sciences, 6(2), 12-15, 2016.
21. [21] Çetin FS, Eryiğit G. “Türkçe hedef tabanli duygu analizi için alt görevlerin incelenmesi-hedef terim, hedef kategori ve duygu sinifi belirleme”. Bilişim Teknolojileri Dergisi, 11(1), 43-56, 2018.
22. [22] Omurca SI, Ekinci E. “Using Adjusted Laplace Smoothing to Extract Implicit Aspects from Turkish Hotel Reviews”. 2018 Innovations in Intelligent Systems and Applications (INISTA), Thessaloniki, Greece, 3-5 July 2018.
23. [23] Karagoz P, Kama B, Ozturk M, Toroslu IH, Canturk D. “A framework for aspect based sentiment analysis on turkish informal texts”. Journal of Intelligent Information Systems, 53(3), 431-451, 2019.
24. [24] Bayraktar K, Yavanoglu U, Ozbilen A. “A Rule-Based Holistic Approach for Turkish Aspect-Based Sentiment Analysis”. 2019 IEEE International Conference on Big Data (Big Data), Los Angeles, USA, 9-12 December 2019.
25. [25] Ekinci E, Omurca Sİ. “Ürün Özelliklerinin Konu Modelleme Yöntemi ile Çıkartılması”. Türkiye Bilişim Vakfı Bilgisayar Bilimleri ve Mühendisliği Dergisi, 9(1), 51-58, 2017.
26. [26] Özyurt B, Akçayol MA. “A new topic modeling based approach for aspect extraction in aspect based sentiment analysis: SS-LDA”. Expert Systems with Applications, 168, 1-14, 2021.
27. [27] Ramos J. “Using Tf-Idf to Determine Word Relevance in Document Queries”. Proceedings of the first instructional conference on machine learning, 242(1), 29-48, 2003.
28. [28] Blei DM, Ng AY, Jordan MI. “Latent dirichlet allocation”. Journal of machine Learning research, 3, 993-1022, 2003.
29. [29] Buenano-Fernandez D, Gonzalez M, Gil D, Lujan-Mora S. “Text Mining of Open-Ended Questions in Self-Assessment of University Teachers: An LDA Topic Modeling Approach”. IEEE Access, 8, 35318-35330, 2020.
30. [30] Lee DD, Seung HS. “Learning the parts of objects by nonnegative matrix factorization”. Nature, 401(6755), 788-791, 1999.
31. [31] Kuang D, Brantingham PJ, Bertozzi AL. “Crime topic modeling”. Crime Science, 6(1), 1-20, 2017.
32. [32] Salur MU, Aydın I. “The impact of preprocessing on classification performance in convolutional neural networks for Turkish text”. 2018 International Conference on Artificial Intelligence and Data Processing (IDAP), Malatya, Turkey, 28-30 September 2018.
33. [33] Uysal AK, Gunal S. “The impact of preprocessing on text classification”. Information Processing & Management, 50(1), 104-112, 2014.
34. [34] Salur MU, Aydin I, Alghrsi SA.“SmartSenti: A twitter-based sentiment analysis system for the smart tourism in Turkey”. 2019 International Conference on Artificial Intelligence and Data Processing Symposium, Malatya, Turkey, 21-22 September 2019.
35. [35] Akın AA, Akın MD. “Zemberek, an open source nlp framework for Turkic languages”. Structure, 10, 1-5, 2007.
36. [36] Pontiki M, Galanis D, Papageorgiou H, Androutsopoulos I, Manandhar S, Mohammad AS, Al-Ayyoub M, Zhao Y, Qin B, De O. “Semeval-2016 task 5: Aspect based sentiment analysis”. Proceedings of the 10th international workshop on semantic evaluation (SemEval-2016), San Diego, California, 16-17 June 2016.