Feature-selection-based dynamic transfer ensemble model for customer churn prediction

Xiao, Jin; Xiao, Yi; Huang, Anqiang; Liu, Dunhu; Wang, Shouyang

doi:10.1007/s10115-013-0722-y

Cited by 45 publications

(23 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even though there are still too many papers with either no attribute selection at all or a selection based only on expert and domain knowledge, the majority of the reviewed publications have used quantitative selection methods based on either statistics or embedded in the own modeling process. Further good news is that the most recent publications of churn analysis in banking seem to be paying adequate attention to this issue; examples of these are the detailed methods that can be found in [24] and, again, in [106]. The following lists provide a non-exhaustive overview of both the more traditional and the CI methods most used for predictive churn modelling.…”

Section: Banking and Financial Servicesmentioning

confidence: 99%

“…Telecommunications: Usage data seem to be the common denominator of recent studies in this business domain [73,106,93,7] and, as seen from the previous paragraph, these type of data are often complemented and enriched using demographics, socio-economic and marketing information.…”

Section: Stage 1: Specific Recommendations Per Industrymentioning

confidence: 99%

“…The obvious recommendation here would be to encourage researchers in the area to take attribute selection seriously as an almost compulsory part of the DM process oriented towards novel knowledge discovery. An example of proper and intensive attribute selection process using CI methods can be found in [106].…”

Section: Stage 2: Specific Recommendations Per Industrymentioning

confidence: 99%

“…They have shown strong and robust prediction performance in many appli-cation areas. According to the combination of algorithms, the most popular ensemble learning methods in the field (although not the only ones; see for instance [106] …”

Section: Stage 3 General Recommendationsmentioning

confidence: 99%

“…In the area of abandonment prediction, the most commonly used modelling techniques, as reflected in the literature, include decision trees (DT), regression analysis [63] and artificial neural networks (ANN) [18], while in more recent years new methods such as support vector machines (SVM) have proven their adequacy [17,13,106].…”

Section: Building Predictive Models Of Abandonmentmentioning

confidence: 99%

See 4 more Smart Citations

Intelligent data analysis approaches to churn as a business problem: a survey

2016

View full text Add to dashboard Cite

Globalization processes and market deregulation policies are rapidly changing the competitive environments of many economic sectors. The appearance of new competitors and technologies leads to an increase in competition and, with it, a growing preoccupation among service providing companies with creating stronger customer bonds. In this context, anticipating the customer's intention to abandon the provider, a phenomenon known as churn, becomes a competitive advantage. Such anticipation can be the result of the correct application of information-based knowledge extraction in the form of business analytics. In particular, the use of intelligent data analysis, or data mining, for the analysis of market surveyed information can be of great assistance to churn management. In this paper, we provide a detailed survey of recent applications of business analytics to churn, with a focus on computational intelligence methods. This is preceded by an in-depth discussion of churn within the context of customer continuity management. The survey is structured according to the stages identified as basic for the building of the predictive models of churn, as well as according to the different types of predictive methods employed and the business areas of their application.Keywords Churn analysis · intelligent data analysis · computational intelligence · customer continuity management · literature survey

show abstract

Section: Banking and Financial Servicesmentioning

confidence: 99%

Section: Stage 1: Specific Recommendations Per Industrymentioning

confidence: 99%

Section: Stage 2: Specific Recommendations Per Industrymentioning

confidence: 99%

Section: Stage 3 General Recommendationsmentioning

confidence: 99%

Section: Building Predictive Models Of Abandonmentmentioning

confidence: 99%

See 3 more Smart Citations

Intelligent data analysis approaches to churn as a business problem: a survey

2016

View full text Add to dashboard Cite

show abstract

Extended artificial immune system–based optimized access control for big data on a cloud environment

Muneeshwari

Athisha

2019

Int J Communication

View full text Add to dashboard Cite

SummaryIntegrating parallel computing and distributed computing together can be obtained by cloud computing (CC). One of the major problems faced by public CC is security regarding data access control. CC permits people to share data, documents, videos, and other types of data. Generally, the cloud data are considered as big data, because the volume of the data is huge and it has a greater number of varieties. In recent days, attribute‐based data sharing applied only for selected data is a crucial problem. One of the existing approaches encrypts data using various kinds of keys based on several types of cryptosystems. However, those kinds of methods have some weaknesses such as inability to handle the attributes effectively, storage of more unwanted copies of the same data, and policy changes. It needs a high amount of computational cost and reduces the efficiency of memory utilization and the computational speed. This paper motivated to design and implement an efficient approach for optimized access control (OAC) for data stored in the cloud to overcome these kinds of issues. The efficiency of the proposed method is proved through a simulation‐based experiment in Cloud Simulator.

show abstract

From Ensemble Learning to Meta-Analytics: A Review on Trends in Business Applications

Haque

Moscato

2019

Business and Consumer Analytics: New Ideas

View full text Add to dashboard Cite

Feature-selection-based dynamic transfer ensemble model for customer churn prediction

Cited by 45 publications

References 34 publications

Intelligent data analysis approaches to churn as a business problem: a survey

Intelligent data analysis approaches to churn as a business problem: a survey

Extended artificial immune system–based optimized access control for big data on a cloud environment

From Ensemble Learning to Meta-Analytics: A Review on Trends in Business Applications

Contact Info

Product

Resources

About