A Novel Customer Churn Model by Markov Chain
Publish Year: 1395
نوع سند: مقاله کنفرانسی
زبان: English
View: 534
This Paper With 10 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
COMCONF04_345
تاریخ نمایه سازی: 10 تیر 1396
Abstract:
Customer churn has been a core research in Customer Relationship Management (CRM) studies in recent years. Customer churn studies contains many aspects, such as modeling churn, estimating churn rate, predicting churn probability, planning strategies to reduce churn. As we know, it is more expensive to gain a new customer than to retain an existing one; therefore companies considerably invest in churn prediction and customer return. Aspects of customer churn are mentioned separately in different studies. Customer churn models are classified into two main groups, always-a-share and lost-for-good. Each of them has pros and cons. In this study we are going to represent a new customer churn model, which benefits the advantages of both classes of churn models. Our model is a Markov chain-based model; and is able to predict customer churn probability. Derived model contains two new definitions of temporal and permanent churn customers. Based on these two concepts and Markov chain analysis the model becomes able to reduce the unnecessary costs paid for customer return. We used the dataset of a manufacturing company to compare the results of our derived model with the normal model of the company, which was an always-a-share one. The results show that our model overcomes the insufficiency of past churn models. The new model is capable to predict different customers churn probability, and is able to select suitable group of customers to concentrate on their return
Keywords:
Authors
Mahsa EsmaeiliGookeh
PhD candidate, IT group Department of Industrial Engineering, K.N. Toosi University of Technology
Mohammad Jafar Tarokh
Professor, IT group Department of Industrial Engineering, K.N. Toosi University of Technology
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :