Resolving Class Imbalance and Feature Selection in Customer Churn Dataset

Hanif, Aamer; Azhar, Noor

doi:10.1109/fit.2017.00022

Cited by 11 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These techniques have been executed and evaluated on random forest classification model. The results illustrate that random oversampling is a better to balanced dataset [9]. Also resolved the unbalanced class problem to predict atrial fibrillation in the obese patient by utilize the SMOTE.…”

Section: Related Workmentioning

confidence: 90%

Toward Constructing a Balanced Intrusion Detection Dataset

Alsameraee¹,

Ibrahem²

2021

sjpas

View full text Add to dashboard Cite

Several Intrusion Detection Systems (IDS) have been proposed in the current decade. Most datasets which associate with intrusion detection dataset suffer from an imbalance class problem. This problem limits the performance of classifier for minority classes. This paper has presented a novel class imbalance processing technology for large scale multiclass dataset, referred to as BMCD. Our algorithm is based on adapting the Synthetic Minority Over-Sampling Technique (SMOTE) with multiclass dataset to improve the detection rate of minority classes while ensuring efficiency. In this work we have been combined five individual CICIDS2017 dataset to create one multiclass dataset which contains several types of attacks. To prove the efficiency of our algorithm, several machine learning algorithms have been applied on combined dataset with and without using BMCD algorithm. The experimental results have concluded that BMCD provides an effective solution to imbalanced intrusion detection and outperforms the state-of-the-art intrusion detection methods.

show abstract

Section: Related Workmentioning

confidence: 90%

Toward Constructing a Balanced Intrusion Detection Dataset

Alsameraee¹,

Ibrahem²

2021

sjpas

View full text Add to dashboard Cite

show abstract

“…With the ensemble methods showing their strength in improving the classification performances, researchers also investigated the combinations of ensemble learning methods -especially bagging and random forests-with resampling methods to tackle the imbalance churn problem. The findings in [28], [29] show that random oversampling combined with random forests yields better results than resampling with random under-sampling and SMOTE. On the other hand, Zhu et al [30] state that bagging and random forests achieve the most promising results with respect to the profit-based measure when there is no resampling involved.…”

Section: Previous Work On Imbalanced Churn Problemmentioning

confidence: 98%

“…Support vector machines (SVM) is a powerful nonparametric supervised learning method based on structural risk minimization [28] and is proven to show good performance especially for binary classification tasks. SVM searches for the optimal separating hyperplane, which is usually a nonlinear decision function 𝑓(𝑥) = (𝑤 𝑇 𝜙(𝑥) + 𝑏), where 𝜙(𝑥) is a nonlinear transform function.…”

Section: Support Vector Machines As a Benchmarkmentioning

confidence: 99%

Müşteri Kaybı Tahmini için Yeniden Örnekleme ve Topluluk Yöntemleri

ÇELİK,

TOLUN TAYALI

2023

Bilişim Teknolojileri Dergisi

View full text Add to dashboard Cite

Churn analysis is a customer relationship management analytics that companies implement to predict the customers who are likely to terminate doing business with them. The success of marketing efforts to retain the existing customers is possible only if probable churners are correctly specified beforehand. Therefore, having powerful models with high prediction capabilities that lead to a profit growth is crucial. The imbalanced nature of churn datasets negatively effects the classification performance of machine learning methods. This study examines resampling –over- and under-sampling- and ensemble learning –bagging, boosting, and stacking– strategies integrated with the cross-validation procedure on imbalanced churn prediction. The experimental results, which are compared to the results of Support Vector Machines taken as the benchmark, show that ensemble methods improve the prediction performances. Also, applying over-sampling achieves a noticeable performance in comparison with the under-sampling approach.

show abstract

“…On the other hand, there is an abundance of classification models proposed for churn prediction, including: Support Vector Machines, Naïve Bayes, Decision Trees and Neural Networks [9]; Support Vector Data Description (SVDD) with random under-sampling and SMOTE oversampling [10]; combinations of random under-sampling and boosting algorithm [11]; random forest combined with random oversampling [12]; Multilayer Perceptron (MLP) neural network [13]; Reverse Nearest Neighborhood and One Class support vector machine (OCSVM) [14]; hybrid combination of well known oversampling technique SMOTE with under-sampling technique [15]; ensemble learning [16] and transfer learning methods [17]. Both complaints and churn are relatively rare events, and building statistical patterns to predict them is extremely difficult due to the imbalance of the data sets: one class (the complaints/churn) is much smaller than the other classes.…”

Section: State Of the Artmentioning

confidence: 99%

Analyzing the Impact of Complaints on Customer Satisfaction in the Travel Industry

Drissi¹,

Ettl²,

Gentile³

et al. 2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Customer satisfaction is crucial for the long term success of any travel service provider. Therefore, identifying situations that can lead to customer dissatisfaction is critical. The strongest evidence of customers dissatisfaction are their complaints. While complaints do not occur very often, they almost always lead to loss of customer goodwill which can cost travel providers millions of dollars in future revenues. In this paper, we describe an approach to proactively identify customers that have the highest propensity to complain as they encounter a travel disruption event. These are invaluable insights that can empower customer service teams with information to deliver a more timely, relevant and impactful service experience. We use three key aspects in this approach: (i) specialized feature engineering for the travel industry; (ii) handling extremely imbalanced data and (iii) adaptation of binary classification, anomaly detection and learning to rank models to our specific task. This research is an important step towards more individualized understanding of customer behavior, and potential service enhancements to further increase customer satisfaction.

show abstract

Resolving Class Imbalance and Feature Selection in Customer Churn Dataset

Cited by 11 publications

References 16 publications

Toward Constructing a Balanced Intrusion Detection Dataset

Toward Constructing a Balanced Intrusion Detection Dataset

Müşteri Kaybı Tahmini için Yeniden Örnekleme ve Topluluk Yöntemleri

Analyzing the Impact of Complaints on Customer Satisfaction in the Travel Industry

Contact Info

Product

Resources

About