Telecom Churn Prediction Research Articles

Customer Churn Prediction in Telecom Based on Machine Learning

Customer churn, a critical metric in the telecommunications industry, reflects the propensity of customers to discontinue service usage, thereby affecting the revenue and growth of a company. Customer churn is greatly affected by individuals and products, and there is no obvious rule, which greatly increases the difficulty of customer loss prediction. With the development of machine learning, the prediction in telecom churn using machine learning has attracted wide attention from researchers. However, there are great differences between different algorithms, which increases the difficulty of application. Therefore, this paper studies the issue of customer churn prediction from the aspect of algorithm. This paper introduces the common data sets and evaluation indexes of customer churn prediction, and then introduces logistic regression, decision tree, K-nearest neighbor, random forest, Gaussian naive Bayes, XGBoost, gradient enhancement and neural network algorithms. Then, through the experimental analysis, the differences between these models are compared. Finally, the paper summarizes the full text and discusses the future research hotspots.

Data mining models in telecom churn prediction

Customer attrition has become one of the biggest problems facing the telecommunications industry due to its fast growth. According to telecom studies, acquiring new customers is more expensive than keeping current ones. Telecom companies may use the information gleaned from telecom data to understand the causes behind customer churn and take steps to maintain their current clientele. This study explores the popular data mining methods for spotting trends in loss of clients. Principal component analysis is used in the survey to reduce the dimension of the attributes. Some conclusions about the relationship between costumer usage data their churn can be summarized through the exploratory data analysis. And three prediction techniques (Logistic Regression, SVM Regression, Random Forest Regression) are applied in the customer churn prediction. This paper compares the accuracy and performance of these models. The result shows, among these models, SVM regression has the highest accuracy. It has an accuracy of 0.92 and a precision of 0.48.

A Comparative Analysis of Serial and Parallel Data Mining Approaches for Customer Churn Prediction in Telecom

A Comparative Analysis of Serial and Parallel Data Mining Approaches for Customer Churn Prediction in Telecom

Telecom Churn Prediction Using Machine Learning

Telecom churn prediction is a critical task for telecom companies to retain their customers. Churn refers to the phenomenon where a customer discontinues their subscription or service with a telecom company. Predicting churn helps telecom companies take proactive measures to prevent churn by identifying potential churners and offering them attractive retention strategies. This abstract presents an overview of the telecom churn prediction problem using machine learning techniques. The telecom churn prediction problem involves analyzing historical customer data, including demographic information, usage patterns, billing details, and service history, to predict whether a customer is likely to churn in the future. Machine learning algorithms are used to learn patterns and relationships from this data and make predictions based on new, unseen data. Telecom churn prediction using machine learning involves preprocessing historical customer data, feature engineering, selecting appropriate machine learning algorithms, evaluating model performance using various metrics, and deploying the best-performing model in a production environment. By implementing this process, telecom companies can reduce churn rates and improve customer satisfaction.

Machine learning based churn prediction in telecom

In the new era of continuous social development, the level of information technology service construction is also increasing, among which the telecommunications service industry has also developed into a relatively large business cluster under this trend. Due to fierce commercial competition, telecommunication companies also face the problem of losing customers. Asking telecom companies to determine the trend of customer churn from the huge amount of customer information data has become a challenge for all telecom companies to solve. This paper takes a telecom company as a reference, using a large amount of subscriber data from it as a research sample, and uses data mining machine learning and other techniques to make churn predictions on customer data, using three classification methods to evaluate the prediction results and help telecom service providers analyze the causes of customer churn. Firstly, the situation faced by the telecom service company is discussed in terms of the relevant background, then by presenting and analyzing the customer data, and finally by using Logical Regression, Mission Tree, and SVM classification models for classification and evaluation, the causes associated with customer churn are summarized, then the methods of customer churn retention are outlined, and the factors associated with customer churn are listed. The factors associated with churn are listed. This provides a reference for an in-depth study of the theoretical approach and application of customer churn.

Research on Application of Stacking Technique in Telecom Churn Prediction

In today's saturated telecommunication market, operators need to fully tap the value of data, strengthen the management of existing users, and reduce customer churns. In order to further to reduce the maintenance cost of enterprises, it is necessary to apply more efficient data mining technology to the prediction of telecommunication user churns in order to improve the operating profit of enterprises. This paper took a telecommunication customer churn data set on Kaggle as the research data, and explores a high accuracy customer churn prediction model. Logistic Regression was exploited to create customer churn prediction models, and the parameters of each model were optimized. On this basis, the Stacking technique was used to fuse the models, and the best combination method is obtained. After research, this paper mainly drew the following conclusions: First, in a single model, the model built with XG Boost algorithm has the highest prediction accuracy, which is 86.20%. Secondly, combining the accuracy rate of training set, cross-validation set and test set, it can be found that XG Boost and Random Forest have the best performance under Stacking fusion algorithm. The accuracy of the above three datasets is 97.61%, 85.84%, 86.51%, respectively, which improves the prediction accuracy of the prediction model.

Telecom Churn Prediction Using Voting Classifier Ensemble Method and Supervised Machine Learning Techniques

In the current fast-paced world, there are a lot of changes and developments in the telecom sector, due to which the telecom companies find themselves in difficulties in retaining the customers who have availed of their services. In order to solve this problem, churn prediction system is needed to predict customer churn. So far, there are many supervised machine learning churn prediction models that compare various machine learning and deep learning models, select one model, and create a whole churn prediction model. The solution proposed has various supervised machine learning models like Support Vector Machine (SVM), Random Forest Classifier, Decision Tree Classifier, and Logistic Regression Classifier and combine all the models together using an ensemble method called Voting Classifier to produce a single model that considers all the mentioned algorithms and produces an optimum result. The above-mentioned model will be trained by the telecom dataset containing the records of 7043 customers, and the target field is classified into churned and stayed. The machine learning algorithm is evaluated using various performance metrics such as the F1 score, precision, confusion matrix, classification report, and accuracy.As the result the churn prediction model has shown 84% accuracy.

Telecom Churn Prediction based on Deep Learning Approach

The transition of customers from one telecom operator to another has a direct impact on the company's growth and revenue. Traditional classification algorithms fail to predict churn effectively. This research introduces a deep learning model for predicting customers planning to leave to another operator. The model works on a high-dimensional large-scale data set. The performance of the model was measured against other classification algorithms, such as Gaussian NB, Random Forrest, and Decision Tree in predicting churn. The evaluation was performed based on accuracy, precision, recall, F-measure, Area Under Curve (AUC), and Receiver Operating Characteristic (ROC) Curve. The proposed deep learning model performs better than other prediction models and achieves a high accuracy rate of 91%. Furthermore, it was noticed that the deep learning model outperforms a small size Neural Network for the customer churn prediction.

Clairvoyant: AdaBoost with Cost-Enabled Cost-Sensitive Classifier for Customer Churn Prediction.

Customer churn prediction is one of the challenging problems and paramount concerns for telecommunication industries. With the increasing number of mobile operators, users can switch from one mobile operator to another if they are unsatisfied with the service. Marketing literature states that it costs 5–10 times more to acquire a new customer than retain an existing one. Hence, effective customer churn management has become a crucial demand for mobile communication operators. Researchers have proposed several classifiers and boosting methods to control customer churn rate, including deep learning (DL) algorithms. However, conventional classification algorithms follow an error-based framework that focuses on improving the classifier's accuracy over cost sensitization. Typical classification algorithms treat misclassification errors equally, which is not applicable in practice. On the contrary, DL algorithms are computationally expensive as well as time-consuming. In this paper, a novel class-dependent cost-sensitive boosting algorithm called AdaBoostWithCost is proposed to reduce the churn cost. This study demonstrates the empirical evaluation of the proposed AdaBoostWithCost algorithm, which consistently outperforms the discrete AdaBoost algorithm concerning telecom churn prediction. The key focus of the AdaBoostWithCost classifier is to reduce false-negative error and the misclassification cost more significantly than the AdaBoost.

Telecom Churn Prediction System Based on Ensemble Learning Using Feature Grouping

In recent years, the telecom market has been very competitive. The cost of retaining existing telecom customers is lower than attracting new customers. It is necessary for a telecom company to understand customer churn through customer relationship management (CRM). Therefore, CRM analyzers are required to predict which customers will churn. This study proposes a customer-churn prediction system that uses an ensemble-learning technique consisting of stacking models and soft voting. Xgboost, Logistic regression, Decision tree, and Naïve Bayes machine-learning algorithms are selected to build a stacking model with two levels, and the three outputs of the second level are used for soft voting. Feature construction of the churn dataset includes equidistant grouping of customer behavior features to expand the space of features and discover latent information from the churn dataset. The original and new churn datasets are analyzed in the stacking ensemble model with four evaluation metrics. The experimental results show that the proposed customer churn predictions have accuracies of 96.12% and 98.09% for the original and new churn datasets, respectively. These results are better than state-of-the-art churn recognition systems.

WITHDRAWN: Heterogeneous ensemble stacking with minority upliftment (HESMU) for churn prediction on imbalanced telecom data

Abstract Telecom industry, being a dynamic and competitive environment inherently exhibits high potential for customer churn. In such a dynamic scenario, conventional performance metrics fail to reflect business goals effectively. This is due to the misalignment existing between the business goals and performance metrics. This research proposes a heterogeneous telecom churn prediction (HESMU) model enhanced by minority elevation strategies, with which data imbalance can be handled effectively and also proposes a loss function that bifurcates loss as a function of unavoidable loss and loss due to the wrong prediction. HESMU is a two-stage process; the initial stage performs first level predictions on the training data using multiple heterogeneous base learners and the second stage performs minority upliftment from the predictions generated by the initial models. The initial stage is composed of Gradient Boosted Trees and Naive Bayes, while the second combiner stage is based on One-Class SVM. Experimental results reveal that the HESMU model shows 1%–7% higher churn prediction levels and 1.3–1.7 times reduced loss levels when compared to existing classifier models and the state of the art churn prediction techniques from literature.

Telecom churn prediction and used techniques, datasets and performance measures: a review

Customer churn prediction in telecommunication industry is a very essential factor to be achieved and it makes direct impact to customer retention and its revenues. Developing a good and effective churn prediction model is very important however it is a time-consuming process. This study presents a very good review of customer churn, its effects, identification of its causes, business needs, methods, and all the techniques used for churn prediction. On the other hand, this study provides the best understanding of the telecom dataset, datasets used by past researches and features used in different researches. Also, this study shows the best techniques used for the churn prediction and describes all performance measures used in the churn prediction models. In this study a wide range of researches are added from the year 2005 to 2020. It includes variety of methods proposed by past researches and technologies used in these researches. At the end, a state of art comparison is added that gives a very good and meaningful comparison of past researches. The study indicates that machine learning techniques are mostly used and feature extraction is a very important task for developing an effective churn prediction model. Deep learning algorithm CNN itself has the capability of feature extraction and establish itself as a powerful technique for churn model, in particular for large datasets. For performance ‘Accuracy’ is a good measure however measuring performance only with ‘Accuracy’ is not sufficient because on small datasets accuracy is more predictable and will be the same. With Accuracy, researchers also need to look at other performance measures such as confusion matrix, ROC, precision. F-measure etc. This study assures that new researchers can find everything regarding their churn prediction model requirements at one place. This study provides a comprehensive view by extensively detailing work which has happened in this area and will act as a rich repositorory of all knowledge regarding churn prediction in telecom sector.

Dynamic behavior based churn prediction in mobile telecom

Customer churn is one of the most challenging problems that affects revenue and customer base in mobile telecom operators. The success of retention campaigns depends not only on the accuracy of predicting potential churners, but with equal importance, it depends on the timing when the prediction is done. Previous works related to churn prediction presented models to predict churn monthly with a focus on the static behavior of customers, and even the studies that considered the dynamic behavior of the customer, looked mainly at the monthly level behavior. However, customer behavior is susceptible to changes over days of month, and during the time leading up to a customer decision to churn, he/she starts behaving differently. Therefore, considering monthly behavioral features negatively affects the predictive performance, because it ignores changes in behavior over days of month. Moreover, predicting churners on monthly basis will be late for customers who decided to leave at the beginning of the month because they will not be detected as churners until the next month. To address these issues, in this paper, we propose daily churn prediction instead of monthly based on the daily dynamic behavior of customer instead of his monthly one. More precisely, we represent customer’s daily behavior as multivariate time series and propose four models to predict churn daily based on this representation. Two models depend on features extracted from the multivariate time series, namely RFM-based model and statistics-based model. While the other models, exploit deep learning techniques for automatic feature extraction, namely LSTM-based model and CNN-based model. The predictive performance of the proposed models were investigated by evaluating them using a 150-day-long dataset collected from MTN operator in the country. The results showed that the daily models significantly outperform the monthly models in terms of predicting churners earlier and more accurately. Furthermore, the LSTM-based model significantly outperforms the CNN-based model. However, the prediction performances of the LSTM-based and the CNN-based models are equal to the prediction performance of the RFM-based model. Moreover, all of these three models significantly outperform the Statistics-based model.

Exploring Hybrid and Ensemble Models for Customer Churn Prediction in Telecom Sector

Most prominent challenges in all business is to retain and satisfy their valuable customers for sustain successfully in the market. Numerous Machine learning approaches are emerging to develop various customer retention models to solve this issue in many applications. This swing is more realized in telecom industry due its enormous significance. This article presents an elaborated survey on machine learning based churn prediction in telecom sector from the year 2000 to 2018. We also extracted the problems and challenges in Telecom Churn Prediction and reported suggestion and solutions. We believe this article helps the researches or data analysts in the telecom field to select optimal and appropriate methods and for designing improved novel model for churn prediction in future.

Customer churn prediction in telecom using machine learning in big data platform

Customer churn is a major problem and one of the most important concerns for large companies. Due to the direct effect on the revenues of the companies, especially in the telecom field, companies are seeking to develop means to predict potential customer to churn. Therefore, finding factors that increase customer churn is important to take necessary actions to reduce this churn. The main contribution of our work is to develop a churn prediction model which assists telecom operators to predict customers who are most likely subject to churn. The model developed in this work uses machine learning techniques on big data platform and builds a new way of features’ engineering and selection. In order to measure the performance of the model, the Area Under Curve (AUC) standard measure is adopted, and the AUC value obtained is 93.3%. Another main contribution is to use customer social network in the prediction model by extracting Social Network Analysis (SNA) features. The use of SNA enhanced the performance of the model from 84 to 93.3% against AUC standard. The model was prepared and tested through Spark environment by working on a large dataset created by transforming big raw data provided by SyriaTel telecom company. The dataset contained all customers’ information over 9 months, and was used to train, test, and evaluate the system at SyriaTel. The model experimented four algorithms: Decision Tree, Random Forest, Gradient Boosted Machine Tree “GBM” and Extreme Gradient Boosting “XGBOOST”. However, the best results were obtained by applying XGBOOST algorithm. This algorithm was used for classification in this churn predictive model.

Performance comparison of six Data mining models for soft churn customer prediction in Telecom

Due to a high competition in the market, the telecom operators are affected by churn, therefore it is very important for them to identify which users are likely to leave them and switch to the competition telecom company. This research uses data on behaviour of the users from telecom systems that serve to identify patterns in behaviours and thereby recognize the churn. Creating new definition of prepaid soft churn based on multiple conditions is valuable contribution of this paper. At preparing data, a selection of useful attributes was made using the Principal Component Analysis (PCA). The normalization of the attribute values has also been made in order to obtain a proper balance of the influence of all the attributes. Common problem with telecom churn prediction data is imbalance, taking into account the target variable. Such a case is also in the data used in this paper, where the percentage of churners is 12%. Comparison of undersampling and oversampling was performed as a method for resolving the data imbalance problem. Data sets with undersampling and oversasmpling have been used to train the decision tree, logistic regression and neural network algorithms and therefore six prediction models for detecting the churn of the Prepaid users in the telecom were created in this paper. Performance analysis and comparison of the six developed Data mining models was also performed.

An enhanced ensemble classifier for telecom churn prediction using cost based uplift modelling

Telecom, being a dynamic and competitive industry which contains an inherently high potential for customer churn, necessitating of accurate churn prediction models. Regular classification approaches fail to effectively predict churn due to low correlation levels between conventional performance metrics and business goals. This work presents an ensemble stacking incorporated with uplifting-based strategies for telecom churn prediction model. Evaluations have been performed based on conventional performance and a cost heuristic, with a major focus upon the cost heuristic. This mode of operation exhibits a high correlation levels between performance indicators and business goals, thus enabling the algorithm suitable for most cost-sensitive applications. A heterogeneous ensemble is created by using multiple algorithms to provide first level predictions. Those predictions with discrepancies are processed at the secondary level using a heuristic based combiner to provide the final predictions. Combination heuristics are fine-tuned based on the cost to predict more accurately concentrating on business goals. Subsequently, Customer uplifting is performed on final predictions, thus making the proposed model 50% more cost efficient than the state-of-the-art ensemble models.

An efficient system for customer churn prediction through particle swarm optimization based feature selection model with simulated annealing

Churn prediction in telecom has gained a huge prominence in the recent times due to the extensive interests exhibited by the stakeholders, large number of competitors and huge revenue losses incurred due to churn. Predicting telecom churn is challenging due to the voluminous and sparse nature of the data. This paper presents a technique for the telecom churn prediction that employs particle swarm optimization (PSO) and proposes three variants of PSO for churn prediction namely, PSO incorporated with feature selection as its pre-processing mechanism, PSO embedded with simulated annealing and finally PSO with a combination of both feature selection and simulated annealing. The proposed classifiers were compared with decision tree, naive bayes, K-nearest neighbor, support vector machine, random forest and three hybrid models to analyze their predictability levels and performance aspects. Accuracy, true positive rate, true negative rate, false positive rate, Precision, F-Measures, receiver operating characteristic and precision-recall plots were used as performance metrics. Experiments reveal that the performance of metaheuristics was more efficient and they also exhibited better predictability levels.

K- local maximum margin feature extraction algorithm for churn prediction in telecom

Telecom customer churn data is not publicly available because involving users' personal privacy. In 2009, the French telecommunications company Orange for knowledge discovery and data mining (KDD) competition provides a telecom customer churn data set KDD Cup 09. In order to solve the high dimensional problem of KDD Cup 09, a new feature reduction method is used to explore the influence of different features on the prediction of classification model. In this paper, a new K- local maximum margin feature extraction algorithm (KLMM) is proposed. Through researching on the diversification subspace partition rules, the corresponding potential field structure is constructed. According to the data source in the dimension of scalability, the intrinsic link between data attributes and classification results is revealed. The extracted features can reduce the dimension of the churn prediction in telecom data. The KLMM method adapts auto selection sigma factor to reflect the anisotropy of features. The potential function is used to assess the weights of attributes and find the potential important weight. Experiments and analysis show that the extracted features by KLMM are more likely to find a classification hyperplane which can separate data points of the different classes.

Churn prediction on huge telecom data using hybrid firefly based classification

Churn prediction in telecom has become a major requirement due to the increase in the number of telecom providers. However due to the hugeness, sparsity and imbalanced nature of the data, churn prediction in telecom has always been a complex task. This paper presents a metaheuristic based churn prediction technique that performs churn prediction on huge telecom data. A hybridized form of Firefly algorithm is used as the classifier. It has been identified that the compute intensive component of the Firefly algorithm is the comparison block, where every firefly is compared with every other firefly to identify the one with the highest light intensity. This component is replaced by Simulated Annealing and the classification process is carried out. Experiments were conducted on the Orange dataset. It was observed that Firefly algorithm works best on churn data and the hybridized Firefly algorithm provides effective and faster results.