Decision Model Development Based on Support Vector Machines Technique (Case Study: Churn Management in Telecommunication Industry)
The telecommunication industry has become an industry with fierce competition. Many service providers offer new products with promising services to regain customers' satisfaction. Therefore, churn management becomes a major concern to monitor customers with high churning probability. By identif...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/12641 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | The telecommunication industry has become an industry with fierce competition. Many service providers offer new products with promising services to regain customers' satisfaction. Therefore, churn management becomes a major concern to monitor customers with high churning probability. By identifying potensial churners, service provider can make decisions to prevent customers shifting to another service provider. This research utilizes Support Vector Machines to predict churners. However, many data mining techniques stop after the data mining final objective has been reached. Decision makers need further input to anticipate churners. The result of SVM can be used to change churners into non-churners. To accommodate domain experts' input, Analytic Hierarchy Process as a multi-criteria decision making model is used to extract actionable decision in terms of customer retention. The model developed is then tested using anonymous data Cell2Cell: The Churn Game. SVM classification gives good accuracy and generalization. Furthermore, AHP gives the best vector as reference to change churners into non-churners. <br />
|
|---|