Likert Tip Veride Bulanık Mantık ve Derin Öğrenme Entegrasyonu

Yıl 2022, , 112 - 125, 28.02.2022

Zeynep Ünal , Emre İpekçi Çetin

https://doi.org/10.35414/akufemubid.1019671

Öz

Derin öğrenme ağları birçok modern uygulamaya sahip olup yüksek performans seviyesi göstermektedir. Derin öğrenme ağlarının gerçek dünyadaki sorunlara uygulamaları yayılmaya devam ederken bunların neden etkili olduğu bilinmemektedir. Ancak deneylerde ağların davranışını inceleyerek bazı yargılarda bulunmak mümkündür. Bu çalışmanın amacı 5 noktalı Likert tipi ölçeğiyle üretilen yapay veri setlerinin üçgensel ya da yamuk bulanık sayılar kullanılarak bulanık forma dönüştürülmesi ve bu yolla verilerin çoğalması durumunda derin öğrenme tekniklerinin performansının analiz edilmesidir. Derin öğrenme ve bulanık mantık tekniklerinin entegrasyonu sonucunda önerilen modelin performansının test edilmesi için memnuniyet tahmin problemi seçilmiştir Bulanık sayılarla oluşturulan veri setleri ile normal veri setinden en az 3 ya da 4 kat daha fazla parametre sayısına ulaşılmaktadır. Böylece büyük veri ile optimizasyon çalışmalarında yerel optimuma tuzağına düşme olasılığı azalmaktadır. Derin öğrenme ile yapılan analizlerde, literatürdeki bulanıklaştırma örneklerine uygun olarak, tepe, maksimum ve minimum değerler için ayrı sonuçlarla durulaştırma gerçekleştirilmiştir. Literatürden farklı olarak bulanık sayıların tek sonuç dizisi üretmesi önerilerek derin öğrenme modelinin performansları araştırılmıştır.

Anahtar Kelimeler

Derin Öğrenme, Lojistik Regresyon, Bulanık Mantık, Likert Ölçeği

Fuzzy Logic and Deep Learning Integration in Likert Type Data

Öz

Deep learning networks have many modern applications and demonstrate a high-performance level. As the applications of deep learning networks to real-world problems continues to spread, the reason why they are effective remains unknown. However, it is possible to make some judgments by examining the behaviour of the network in experiments. The main aim of this study is to analyse the performance of deep learning techniques in the form of a 5-point Likert-type scale by converting the artificial data sets into a fuzzy form using triangular or trapezium fuzzy numbers. To test the performance of the proposed model, which is the integration of deep learning and fuzzy logic techniques, the satisfaction estimation problem was chosen. Data sets consisting of fuzzy numbers which reach at least three or four times more parameters than normal data sets. Thus, it decreases the possibility of falling into the local optimum trap in optimization studies with big data. In the analysis conducted with deep learning, in accordance with the fuzzification examples in the literature, the defuzzification was carried out with separate results for peak, maximum, and minimum values. In contrast to the literature, the performances of the deep learning model were investigated by suggesting that fuzzy numbers produce a single result series.

Anahtar Kelimeler

Deep Learning, Logistic Regression, Fuzzy Logic, Likert Scale

Kaynakça

• Albon, C., 2018. Machine Learning with Python Cookbook, USA: O’Reilly Media, Inc, 180-186.
• Araç, Y. E., Gürhanlı, A., 2020. Yapay sinir ağını kullanarak müşteri memnuniyeti analizi. Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi, 11(1), 39-55.
• Bahadır, E. (2017). Bulanık Mantık Yaklaşımının Eğitim Çalışmalarında Kullanılmasının Alan Yazın Işığında Değerlendirilmesi. Uluslararası Sosyal ve Eğitim Bilimleri Dergisi, 4(7), 28-42.
• Bekiros, S., Loukeris, N., Matsatsinis, N., Bezzina, F., 2019. Customer satisfaction prediction in the shipping industry with hybrid meta-heuristic approaches. Computational Economics, 54(2), 647-667.
• Biyan, M., & Bircan, H. (2018). Daha önce geliştirilmiş likert tipi bir ölçek ile tip-1 ve tip-2 bulanık likert ölçeğinin sonuçlarının karşılaştırılması. Uluslararası İktisadi ve İdari İncelemeler Dergisi, 369-382.
• Cong, P., Wang, C., Ren, Z., Wang, H., Wang, Y., Feng, J.,N. 2016. Unsatisfied customer call detection with deep learning, In 2016 10th International Symposium on Chinese Spoken Language Processing (ISCSLP), Tianjin, China.
• Deng, W. J., Pei, W., 2009. Fuzzy neural based importance-performance analysis for determining critical service attributes. Expert Systems with Applications, 36(2), 3774-3784.
• Deng, L., Yu, D., 2014. Deep learning: methods and applications. Foundations and Trends in Signal Processing, 7(3-4), 197-387.
• El Hatri, C., Boumhidi, J., 2018. Fuzzy deep learning based urban traffic incident detection. Cognitive Systems Research, 50, 206-213.
• Feng, S., Zhou, H., Dong, H., 2019. Using deep neural network with small dataset to predict material defects. Materials & Design, 162,300-310.
• Güner, N., Çomak, E., 2014. Lise öğrencilerinin matematik dersine yönelik tutumlarının bulanık mantık yöntemi ile incelenmesi. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 20(5), 189-196.
• Goodfellow, I., Bengio, Y., Courville, A., 2016. Deep learning, USA: MIT Press, 130-135
• Hahn, S., Choi, H., 2020. Understanding dropout as an optimization trick. Neurocomputing, 398, 64-70.
• Hendalianpour, A., Razmi, J., 2017. Customer satisfaction measurement using fuzzy neural network. Decision Science Letters, 6(2), 193-206.
• Heaton, J. (2015). Artificial Intelligence for Humans, Volume 3: Neural Networks and Deep Learning. Heaton Research Inc, Chesterfield, ABD, 30-55.
• Hinton, G., Osindero, S., Teh, Y. W., 2006. A fast-learning algorithm for deep belief nets. Neural computation, 18(7), 1527-1554.
• Ishibuchi, H., Nii, M., 1998. Fuzzification of input vectors for improving the generalization ability of neural networks. In 1998 IEEE International Conference on Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence, Anchorage, AK, USA, 1153-1158.
• Islam, K. K., and R. G. Raj, 2017. "Real-time (vision-based) road sign recognition using an artificial neural network." Sensors, 17(4), 853.
• Jahandideh, S., Asefzadeh, S., Jahandideh, M., Asadabadi, E. B., Jafari, A., 2013. The comparison of methods for measuring quality of hospital services by using neural networks: A case study in Iran. International Journal of Healthcare Management, 6(1), 45-50.
• Kappor, R., Walters, S. P., Al-Aswad, L. A., 2018. The current state of artificial intelligence in ophthalmology. Survey of Ophthalmology, 64(2), 233-240.
• Kalinić, Z., Marinković, V., Djordjevic, A., Liebana-Cabanillas, F., 2019. What drives customer satisfaction and word of mouth in mobile commerce services? A UTAUT2-based analytical approach. Journal of Enterprise Information Management, 33(1), 71-94.
• Kennedy, K., Delany, S. J., Mac Namee, B., 2011. A Framework for Generating Data to Simulate Application Scoring. In Credit Scoring and Credit Control XII, Conference Proceedings, Edinburg.
• Li, Q, 2013. A novel Likert scale based on fuzzy sets theory. Expert Systems with Applications, 40(5), 1609-1618.
• Lin, Y. S., 2017. Causal complexity for passengers’ intentions to re-ride. Quality & Quantity, 51(5), 1925-1937.
• Mahani, A., Baba Ali, A. R., 2020. Classification problem in imbalanced datasets. In Recent Trends in Computational Intelligence, eds A. Sadollah, and T. Sinha, London: IntechOpen Press, 1-23.
• Najmi, A., Kanapathy, K., Aziz, A. A., 2021. Understanding consumer participation in managing ICT waste: Findings from two-staged Structural Equation Modeling-Artificial Neural Network approach. Environmental Science and Pollution Research, 28(12), 14782–14796.
• Subroto, A., Christianis, M., 2021. Rating prediction of peer-to-peer accommodation through attributes and topics from customer review. Journal of Big Data, 8(1), 1-29.
• Tabrizi, T. S., Khoie, M. R., Sahebkar, E., Rahimi, S., Marhamati, N., 2016. Towards a patient satisfaction-based hospital recommendation system”, In 2016 International Joint Conference on Neural Networks, Vancouver, Canada, 131-138.
• Tóth, Z. E., Árva, G., Dénes, R. V., 2020. Are the ‘Illnesses’ of Traditional Likert Scales Treatable? Quality Innovation Prosperity, 24(2), 120-136.
• Tsaur, S. H., Chiu, Y. C., Huang, C. H., 2002. Determinants of guest loyalty to international tourist hotels - a neural network approach”. Tourism Management, 23(4), 397-405.
• Tóth, Z. E., Jónás, T., Dénes, R. V., 2019. Applying flexible fuzzy numbers for evaluating service features in healthcare–patients and employees in the focus. Total Quality Management & Business Excellence, 30(sup1), 240-254.
• Wahyudi, R. D., Hadiyat, M. A., Hartono, M., 2018. Predicting Service Reliability-Using Survival Analysis of Customer Fuzzy Satisfaction. The Asian Journal of Technology Management, 11(2), 79-93.
• Patterson, J., Gibson, A., 2017. Deep Learning: A Practitioner's Approach”, USA: O'Reilly Media, Inc, 70-77.
• Peng, T., Hanke, F., 2016. Towards a Synthetic Data Generator for Matching Decision Trees, In Proceedings of the 18th International Conference on Enterprise Information Systems (ICEIS), 135-141.
• Raschka, S., Mirjalili, V., 2017. Machine Learning and Deep Learning with Python, scikit-learn and TensorFlow, UK: Packt Publishing, 189-195.
• Sreekumar, S., Mahapatra, S., 2015. Service quality of Indian banks: A fuzzy inference system approach. Asian Academy of Management Journal, 20(2), 9-80.
• Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R., 2014. Dropout: a simple way to prevent neural networks from overfitting. The Journal of Machine Learning Research, 15(1), 1929-1958.
• Wang, H., Xu, Z., Pedrycz, W., 2017. An overview on the roles of fuzzy set techniques in big data processing: Trends, challenges and opportunities. Knowledge-Based Systems, 118, 15-30.
• Wang , W. M., Wang, J. W., Barenji, A. V., Li, Z., & Tsui, E., 2019. Modeling of individual customer delivery satisfaction: an AutoML and multi-agent system approach. Industrial Management & Data Systems, 19(4), 840-866.
• Wright, J. L., Manic, M., 2010. Neural network architecture selection analysis with application to cryptography location, In The 2010 International Joint Conference on Neural Networks (IJCNN) IEEE, Barcelona, Spain.
• Yau, H. K., Tang, H. Y. H., 2018. Analyzing customer satisfaction in self-service technology adopted in airports. Journal of Marketing Analytics, 6(1), 6-18.
• Zheng, H., Yang, Z., Liu, W., Liang, J., Li, Y., 2015. Improving deep neural networks using softplus units, In 2015 International Joint Conference on Neural Networks (IJCNN) IEEE, Killarney, Ireland.