A Machine Learning Approach to Classify the Telecommunication Customers Based on Their Profitability

A Machine Learning Approach to Classify the Telecommunication Customers Based on Their Profitability

Selvaratnam Mishoba, Kuhaneswaran Banujan, Senthan Prasanth, B. T. G. S. Kumara
DOI: 10.4018/978-1-7998-9553-4.ch001
(Individual Chapters)
No Current Special Offers


Customer profitability is one of the most critical problems faced by businesses today. Keeping an existing customer is more valuable than gaining a new subscriber in the telecommunication industry. As a result, anticipating customer attrition behavior in advance is challenging. This behavior has prompted most researchers to establish a model for categorizing clients based on their profitability levels in various businesses. This study was carried out with the assistance of a local telecommunication service provider. Approximately 10,000 pre-paid subscriber details with 12 attributes were acquired. Furthermore, the classification technique was used to reduce the dimensionality between features and classify the high profitable customers, low profitable customers, and average profitable customers. The data was then fed into various supervised learning algorithms to choose the optimum algorithm by considering certain evaluation metrics for developing the final prediction model. The proposed approach revealed that the SVM outperformed all other techniques with greater accuracy of 80.00%.
Chapter Preview


A key strategic objective of many companies has been to find ways to improve customer satisfaction that directly impact on company’s income and revenue. Companies can obtain customer profitability from serving a customer or customer base for a specific period. Understanding customers’ profitability is the most critical factor for an organization to make strategic decisions to make the company a stable position and maintain its reputation (Mulhern, 1999).

Information and communication have significantly changed smartphones instead of fixed-line telephones. The communication medium changed from traditional calls and text messages to alternatives such as Skype, Zoom, Google Duo, and social media. Today’s status of the telecommunication industry in the national economy continues to improve, influencing economic growth plays a crucial role in increasing a country’s economic development (Reza, Nahar, & Akter, n.d.).

The telecommunication industry plays a vital role in the fast-moving modern world. It delivers the customers to the communication medium for a set of voice, SMS, and data. Telecommunication uses electrical devices such as smartphones, microwave communications, and the internet in the contemporary world. At the same time, the telecommunication industry is highly competitive because multiple telecommunication service providers provide various solutions to the customers; thus, customers move from one service provider to another depending on their needs quickly (Sujah & Rathnayaka, 2019).

Making better customer relationship management is the way to manage the interaction with current and future customers to increase profitability. Most successive organizations continuously research their customers in various aspects such as customer identification, customer attraction, customer retention, and customer development to achieve a high level of customer relationship. Maintaining an efficient relationship with customers can directly affect the industry’s profit (Arumawadu, Rathnayaka, & Illangarathne, 2015).

The telecommunication industry is under revenue threats of losing potential customers. Quality of the service or product, the price, and the quality of customer service are the three factors involved in considering the customer’s availability in an industry. As a solution, the telecommunication industry needs to increase its capability to understand customer needs and preferences to stay competitive with other sectors, achieve a high level of customer profitability, and continue revenue growth for a long time (Anand, Brunner, Ikegwu, & Sougiannis, 2019).

Customers play a vital role in business success to increase profitability. The customers are categorized into different groups based on their behavioral patterns, such as low, average, and high profitability customers (Sujah & Rathnayaka, 2019). The customer providing the highest profit to the industry will be considered a High Profitable Customer (Chang & Chong, 2011). Providing different services to different groups of customers based on their profitability level increases customer satisfaction to increase the profitability of the telecommunication industry (Xu, Qiu, & Qiu, 2003).

Data Mining (DM) is extracting hidden information from a large amount of data used to discover new, accurate and valuable patterns (Liang, 2010). DM is a logical process that most researchers use to search through many data to find helpful information for the organization or individual who needs it. Commonly used DM technique has various algorithms and techniques like Classification, Clustering, Regression, Artificial Intelligence, Neural Networks, Associated Rules, Decision Trees, Genetic Algorithm, the Nearest Neighbor method, etc., are used for knowledge discovery from the comprehensive dataset (Han, Pei, & Kamber, 2011).

In this study, we planned to produce an efficient way to predict the profitability of a particular customer or group of customers to the better decision for the future. It isn’t easy with the traditional methods and tools (Ćamilović, 2008). For this problem, this study has proposed a new methodology as a solution using DM techniques and machine learning algorithms to get the most precise results. In this research study, the customers are divided into three main categories: (i) Low profitable customers, (ii) Average Profitable Customers, (iii) High profitable customers (Gašpar, Markić, & Ćorić, 2012). Machine learning was adapted to perform the aforementioned classification efficiently.

Key Terms in this Chapter

Supervised Machine Learning: Supervised learning, often known as supervised machine learning, is artificial intelligence and machine learning subcategory. It uses labelled datasets to train algorithms that accurately classify data or predict outcomes defined it. As input data is fed into the model, the weights are adjusted until the model is correctly fitted during the cross-validation phase.

Machine Learning: Machine learning (ML) is a sort of artificial intelligence (AI) that allows the software to improve its accuracy at predicting outcomes without being explicitly programmed to do so.

Prediction: When anticipating the likelihood of a given result, such as whether or not a customer would churn in 30 days, “prediction” refers to the output of an algorithm after it has been trained on a previous dataset and applied to new data.

Cross-Validation: Cross-validation is a resampling method for evaluating machine learning models on a small sample of data.

Multi-Class Classification Problem: A classification problem with more than two classes, such as identifying a pug, bulldog, or trabeation mastiff from a series of dog breed images. Multi-class classification assumes that each sample is assigned to one of several classes; for example, a dog can be a pug or a bulldog, but not both at the same time.

Complete Chapter List

Search this Book: