Citation

Maw, Maw and Haw, Su Cheng and Ng, Kok Why (2022) An Enhanced Customer Churn Prediction Framework using Data Mining and Machine learning Techniques. In: Postgraduate Colloquium December 2022, 1-15 December 2022, Multimedia University, Malaysia. (Unpublished)

Text 17.MAW MAW.pdf - Submitted Version Restricted to Registered users only Download (579kB)

Abstract

Decision-making systems (DMS) have been transformed from human-based into machine-based decision-making systems (MBDMS) and are bringing tremendous success in making fast and proper decisions. In 2016, it was shockingly found out that MBDMS produced the biased results toward a certain group of people through a criminal justice assessment used in US courts (Angwin et. al, 2016). This fact alarms the AI researchers to put more focus on algorithmic fairness (AF) in machine-learning based applications in various domains. In customer-oriented and profit-centered sectors, customer churn prediction (CCP) is one of the most pervasive applications in telecom business, which forecasts the customers who are likely to stop their services in the near future (Kim et al., 2022). The nature of CCP task requires to use the historical data in the prediction process, which would probably contaminate with biased data. In addition, churn datasets are originally unbalanced due to the massive availability of data but contain a few examples of churners. Thus, data sampling techniques (DTSs) are usually required to be applied in the preprocessing phase.

Item Type:	Conference or Workshop Item (Poster)
Uncontrolled Keywords:	Framework, algorithmic
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions:	Faculty of Computing and Informatics (FCI)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	19 Dec 2022 09:12
Last Modified:	19 Dec 2022 09:12
URII:	http://shdl.mmu.edu.my/id/eprint/10932

Downloads

Edit (login required)