Fuzzy Classification of Customer Insolvency in Mobile Telecommunication

Fuzzy Classification of Customer Insolvency in Mobile Telecommunication

Walid Moudani, Grace Zaarour, Félix Mora-Camino
Copyright: © 2014 |Pages: 29
DOI: 10.4018/ijdsst.2014070101
(Individual Articles)
No Current Special Offers


This paper proposes a predictive model to handle customer insolvency in advance for large mobile telecommunication companies for the purpose of minimizing their losses while preserving an overall satisfaction of the customers which may have important consequences on the quality and on the consume return of the operations. A new mathematical formulation taking into consideration a set of business rules and the satisfaction of the customers is proposed. However, the customer insolvency is defined to be a classification problem since our main purpose is to categorize the customer in one of the two classes: potentially insolvent or potentially solvent. Therefore, a model with precise business prediction using the knowledge discovery and Data Mining techniques on an enormous heterogeneous and noisy data is proposed. A fuzzy approach to evaluate and analyze the customer behavior leading to segment them into groups that provide better understanding of customers is developed. These groups with many other significant variables feed into a classification algorithm based on Rough fuzzy Sets technique to classify the customers. A real case study is considered here, followed by analysis and comparison of the results for the reason to select the best classification model that maximizes the accuracy for insolvent customers and minimizes the error rate in the misclassification of solvent customers.
Article Preview

1. Introduction

Telecommunication Company worldwide suffers from customers who use the provided services without paying. The estimated losses amount to several billions of dollars in uncollectible debt per day. Even though this is a small percentage comparing to the Telecom Operators’ revenue, it is still a significant loss. The mobile telecommunication industry stores and generates tremendous amounts of raw and heterogeneous data that provides rich fields for analysis. This data includes Event Details describing the calls, SMSes, MMSes, and other events that traverse the Communication Network Switches, Network Elements Data which describes the state of the hardware and software components, and Customer Data that illustrates the different profiles of the Telecom Operators’ customers, the services provided to each customer, and the financial transactions of the customers with the Operators. Moreover, these companies apply late precautions against the insolvent customers that result in no considerable effect. Thus, detection and prevention from the insolvency is one of the main objectives of the telecommunication industry.

However, the volume of data being generated nowadays is increasing at phenomenal rate. So, extracting useful knowledge from such data collections is an important and challenging issue. In order to build such a non-trivial model, many researches were carried out on the feasibility of using the Data Mining techniques which comes from the need of analyzing high volumes of data collected by the telecommunication companies and related to different kinds of transactions between the company and its customers. Examples of such data are:

  • Customer Data: Include the customers’ profiles like names and address information, service plans and contact information, and the financial transactions of the customers with the company, credit scores, installments and payment history. Thus Customer data is typically used to supplement call detail data in our study

  • The Unbilled Calls: Illustrate in details the customer usage of service offered like the local calls, the international calls, the SMSes, The MMSes, the free airtime packages in real-time.

Data mining refers to extracting knowledge from large amounts of data. Data mining treats as synonym for another popularly used term, Knowledge Discovery in Databases (KDD) (Fayyad, Piatetsky-shapiro, Smyth & Widener, 1996). KDD consists of the following steps to process it such as: data cleaning, data integration, data selection, data transformation, data mining, pattern evaluation and Knowledge presentation. KDD is the nontrivial process of identifying valid, novel, potentially useful and ultimately understandable patterns in data. Data mining is not a single technique, some commonly used techniques are: Statistical Methods, Case-Based Reasoning (CBR), Neural Networks, Decision Trees, Rule Induction, Bayesian Belief Networks (BBN), Genetic Algorithms, Fuzzy Sets and Rough Sets (Vercellis, 2009; Meier, Nicolas, Albrecht & Sarakinos, 2005; Archer & Wang, 1993; Estévez, Held & Perez, 2006; Burge & Shawe-Taylor, 2001; Burge, Shawe-Taylor, Cooke, Moreau, Preneel & Stoermann, 1997; Hollmen, 2000; Pinheiro, Evsukoff & Ebecken, 2006). Data Mining have been widely recognized as powerful tool for fraud detection and direct marketing applications. Nevertheless, one of the important issues in Data Mining is Data Mining deployment actions, i.e. the results of feedback actions taken from knowledge extracted from data. Moreover, a clear business understanding and the analysis of goals are essentials.

Complete Article List

Search this Journal:
Volume 16: 1 Issue (2024)
Volume 15: 2 Issues (2023)
Volume 14: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 13: 4 Issues (2021)
Volume 12: 4 Issues (2020)
Volume 11: 4 Issues (2019)
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing