Açıklanabilir Yapay Zekaya Dayalı Müşteri Kaybı Analizi ve Elde Tutma Önerisi

Yıl 2024, , 13 - 23, 30.04.2024

İbrahim Kök

https://doi.org/10.46387/bjesr.1344414

Öz

Mobil telekomünikasyon pazarında aboneler yüksek hizmet kalitesi, rekabetçi fiyatlandırma ve gelişmiş servis beklentisindedirler. Müşteri bu beklentilerini telekom servis sağlayıcısından karşılayamaması durumunda onu değiştirme yoluna gitmektedir. Hizmet sağlayıcı operatörlerin ise abone kaybı olarak nitelendirilen bu durumla başa çıkmak için abonelerin iletişim kalıpları, davranışları ve abonelik planlarına ait verileri analiz ederek stratejik öngörü sağlayan yorumlanabilir müşteri kaybı tahmin modellerine ihtiyacı vardır. Bu çalışmada biz K-En Yakın Komşu, Karar Ağacı, Rastgele Orman, Destek Vektör Makinesi ve Naïve Bayes algoritmalarına dayalı müşteri kaybı tahmin modelleri geliştiriyoruz. Aynı zamanda en başarılı algoritma sonuçlarının açıklanabilirliği ve yorumlanabilirliği için ELI5, LIME, SHAP ve karşıolgusal açıklanabilir yapay zeka yöntemleri kullanıyoruz. Bu sayede geliştirilen modeller incelenen abonelerin sadece operatörü değiştirip değiştirmediği değil aynı zamanda abone davranışına sebep olan özellikleri de çıktı olarak vermektedir. Geliştirilen açıklanabilir modeller aracılığıyla servis sağlayıcılara müşteri davranışlarının nasıl ve neden gerçekleştiğine dair kapsamlı analizler sunuyoruz.

Anahtar Kelimeler

Müşteri Kaybı Analizi, Makine Öğrenmesi, Açıklanabilir Yapay Zeka, Abone Elde Tutma

Explainable AI-Driven Churn Analysis And Retention Recommendation

Öz

In the mobile telecommunications market, subscribers expect high service quality, competitive pricing and improved service. If the customer is unable to meet these expectations from the telecom service provider, he/she switches. In order to cope with this situation, service provider operators need interpretable churn prediction models that provide strategic insights by analyzing data on subscribers' communication patterns, behaviors and subscription plans. In this paper, we develop churn prediction models based on K-Nearest Neighbor, Decision Tree, Random Forest, Support Vector Machine and Naïve Bayes algorithms. We also use ELI5, LIME, SHAP and Counterfactual explainable artificial intelligence methods for the explainability and interpretability of the most successful algorithm results. In this way, the developed model outputs not only whether the examined subscribers change the operator or not, but also the features that cause the subscriber behavior. Through the developed explainable models, we provide service providers with comprehensive analyses of how and why customer behavior occurs.

Anahtar Kelimeler

Churn Analysis, Machine Learning, Explainable Artificial Intelligence, Subscriber Retention

Kaynakça

• P. Taylor, “Number of smartphone mobile network subscriptions worldwide from 2016 to 2022, with forecasts from 2023 to 2028”i 2023. www.statista.com/statistics/330695/number-of-smartphone-users-worldwide/
• K. Coussement, S. Lessmann, and G. Verstraeten, “A comparative analysis of data preparation algorithms for customer churn prediction: A case study in the telecommunication industry”, Decis Support Syst, vol. 95, pp. 27-36, 2017.
• T. Xu, Y. Ma, and K. Kim, “Telecom churn prediction system based on ensemble learning using feature grouping,” Applied Sciences, vol. 11, no. 11, p. 4742, 2021.
• The European Business Review, “How costly is customer churn in the telecom industry, the European business review”, [Çevrimiçi]. Erişim:https://www.europeanbusinessreview.com/how-costly-is-customer-churn-in-the-telecom-industry/
• E. Yeboah-Asiamah, B. Narteh, and M.A. Mahmoud, “Preventing Customer Churn in the Mobile Telecommunication Industry: Is Mobile Money Usage the Missing Link?”, (in English), J Afr Bus, vol. 19, no. 2, pp. 174-194, 2018.
• H. Li et al., “Enhancing Telco Service Quality with Big Data Enabled Churn Analysis: Infrastructure, Model, and Deployment,” (in English), J Comput Sci Tech-Ch, vol. 30, no. 6, pp. 1201-1214, Nov 2015.
• H. Li, D. L. Yang, L. L. Yang, Y. Lu, and X. L. Lin, “Supervised Massive Data Analysis for Telecommunication Customer Churn Prediction,” (in English), Proceedings of 2016 Ieee International Conferences on Big Data and Cloud Computing (Bdcloud 2016) Social Computing and Networking (Socialcom 2016) Sustainable Computing and Communications (Sustaincom 2016) (Bdcloud-Socialcom-Sustaincom 2016), pp. 163-169, 2016.
• M.Z. Kastouni and A. Ait Lahcen, “Big data analytics in telecommunications: Governance, architecture and use cases,” Journal of King Saud University - Computer and Information Sciences, vol. 34, no. 6, pp. 2758-2770, 2022.
• S. Ali et al., “Explainable Artificial Intelligence (XAI): What we know and what is left to attain Trustworthy Artificial Intelligence,” Inform Fusion, vol. 99, p. 101805, 2023.
• A. Barredo Arrieta et al., “Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI,” Inform Fusion, vol. 58, pp. 82-115, 2020/06/01/ 2020.
• İ. Kök, Y.F. Okay, Ö. Muyanlı, and S. Özdemir, “Explainable Artificial Intelligence (XAI) for Internet of Things: A Survey,” Ieee Internet Things, vol. 10, no. 16, pp. 14764-14779, 2023.
• P. Kisioglu and Y.I. Topcu, “Applying Bayesian Belief Network approach to customer churn analysis: A case study on the telecom industry of Turkey,” (in English), Expert Syst Appl, vol. 38, no. 6, pp. 7151-7157, Jun 2011.
• N. Lu, H. Lin, J. Lu, and G. Zhang, “A customer churn prediction model in telecom industry using boosting,” Ieee T Ind Inform, vol. 10, no. 2, pp. 1659-1665, 2012.
• H. Jung, J. Mo, and J. Park, “A Data-Driven Customer Quality of Experience System for a Cellular Network,” (in English), Mob Inf Syst, vol. 2017, 2017.
• M. Bagri, J.K. Singh, M.K. Abhilash, R.S. Sunitha, and S. Kumar, “Churn Analysis in Telecommunication Industry,” (in English), 2018 International Conference on Automation and Computational Engineering (Icace), pp. 126-132, 2018.
• S. Mitrović, B. Baesens, W. Lemahieu, and J. De Weerdt, “On the operational efficiency of different feature types for telco Churn prediction,” European Journal of Operational Research, vol. 267, no. 3, pp. 1141-1155, 2018.
• A. De Caigny, K. Coussement, and K.W. De Bock, “A new hybrid classification algorithm for customer churn prediction based on logistic regression and decision trees,” European Journal of Operational Research, vol. 269, no. 2, pp. 760-772, 2018.
• T.W. Cenggoro, R.A. Wirastari, E. Rudianto, M.I. Mohadi, D. Ratj, and B. Pardamean, “Deep learning as a vector embedding model for customer churn,” Procedia Computer Science, vol. 179, pp. 624-631, 2021.
• S.L. Wu, W.C. Yau, T.S. Ong, and S.C. Chong, “Integrated Churn Prediction and Customer Segmentation Framework for Telco Business,” (in English), Ieee Access, vol. 9, pp. 62118-62136, 2021.
• S.M. Shrestha and A. Shakya, “A Customer Churn Prediction Model using XGBoost for the Telecommunication Industry in Nepal,” Procedia Computer Science, vol. 215, pp. 652-661, 2022.
• A. Amin, A. Adnan, and S. Anwar, “An adaptive learning approach for customer churn prediction in the telecommunication industry using evolutionary computation and Naïve Bayes,” Applied Soft Computing, vol. 137, 2023.
• L. Saha, H.K. Tripathy, T. Gaber, H. El-Gohary, and E.-S. M. El-kenawy, “Deep churn prediction method for telecommunication industry,” Sustainability-Basel, vol. 15, no. 5, p. 4543, 2023.
• B. Prabadevi, R. Shalini, and B.R. Kavitha, “Customer churning analysis using machine learning algorithms,” International Journal of Intelligent Networks, vol. 4, pp. 145-154, 2023.