repository
Special Graduate Schools
Graduate School of Science and Technology
Department of Industrial Engineering
3. Theses(Master)
CRM 고객반응 모델링을 위한 그래프 앙상블 알고리즘
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | 신현정 | - |
| dc.contributor.author | 최인애 | - |
| dc.date.accessioned | 2018-11-08T07:50:19Z | - |
| dc.date.available | 2018-11-08T07:50:19Z | - |
| dc.date.issued | 2009-02 | - |
| dc.identifier.other | 9756 | - |
| dc.identifier.uri | https://dspace.ajou.ac.kr/handle/2018.oak/7742 | - |
| dc.description | 학위논문(석사)--아주대학교 일반대학원 :산업공학과,2009. 2 | - |
| dc.description.tableofcontents | 1 Introduction 1 2 Literature Review 5 2.1 Graph-based Semi-supervised Learning 5 2.2 Graph Sharpening 7 2.3 Ensemble Learning 9 3 Sharpened Graph Ensemble 12 4 Experiments 14 4.1 Artificial Data 14 4.2 Benchmark Data 17 4.3 Response Modeling for Insurance Company 21 5 Conclusion 27 Bibliography 28 |1. An ordinary graph by W: The labeled node is denoted as "+1" or "-1", and the unlabeled node as "?". The edge has no directionality. 7 2. A sharpened graph by Ws: By graph sharpening some edges have been removed or have assumed directionality according to the importance of the information flow 8 3. Diversification of member in ensemble learning 11 4. The procedure of the proposed method-sharpened graph ensemble 13 5. Two-moon toy problem 15 6. The change in the AUC over hyperparameter variation (k and µ) 16 7. The AUC comparison for two methods in three different cases. More dots in upper diagonal half indicate that the method in vertical axis outperforms the other assigned in horizontal axis. 19 8. The AUC comparison of Single-Original and Ensemble-Sharpened 20 9. The comparison of Single-Original and Single-Sharpened (All, Demographic) 23 10. The box plots show the distribution of Single-Original (All, Demographic, Monetary, Frequency)and Ensemble-Sharpened (proposed method) 25|1. Summary of the five benchmark data sets 17 2. AUC comparison for the five benchmark data sets (mean ± std) 18 3. The summary of the CoIL data sets 22 4. The comparison of classification model 26 | - |
| dc.language.iso | eng | - |
| dc.publisher | The Graduate School, Ajou University | - |
| dc.rights | 아주대학교 논문은 저작권에 의해 보호받습니다. | - |
| dc.title | CRM 고객반응 모델링을 위한 그래프 앙상블 알고리즘 | - |
| dc.title.alternative | Choi Inae | - |
| dc.type | Thesis | - |
| dc.contributor.affiliation | 아주대학교 일반대학원 | - |
| dc.contributor.alternativeName | Choi Inae | - |
| dc.contributor.department | 일반대학원 산업공학과 | - |
| dc.date.awarded | 2009. 2 | - |
| dc.description.degree | Master | - |
| dc.identifier.localId | 567755 | - |
| dc.identifier.url | http://dcoll.ajou.ac.kr:9080/dcollection/jsp/common/DcLoOrgPer.jsp?sItemId=000000009756 | - |
| dc.subject.keyword | Response modeling | - |
| dc.subject.keyword | Semi-supervised learning | - |
| dc.subject.keyword | Graph sharpening | - |
| dc.subject.keyword | Ensemble learning | - |
| dc.description.alternativeAbstract | Response modeling in the domain of customer relationship management (CRM) is to predict certain customers who are more likely to buy a promoted product or service based on the customer database. The customers who have been targeted in a preliminary campaign have the labels of either respondent or non-respondent and thus are called labeled, whereas the others, who have been excluded from the campaign and thus do not have labels, are called unlabeled. Because of the limited marketing cost for the campaign, only a small part of the customer data is labeled whereas the rest large amount of the customer data remains unlabeled. In the light of that, the most recently proposed semi-supervised learning (SSL) is well-suited for response modeling. In the framework of SSL, the labeled data points are used for the basic modeling and the unlabeled data points are also used for exploiting the information on the underlying manifold structure of input space. However, there are some technical difficulties. Because of the lack of the labeled data points, it becomes difficult to find the optimal value of the learning hyperparameter and to figure out the influence of noise on performance. To circumvent the addressed difficulties, we propose to employ ensemble learning and graph sharpening. The ensemble learning replaces the hyperparamer selection procedure to an ensemble network of the committee trained with various values of hyperparameter. On the other hand, graph sharpening helps to remove unhelpful information caused by noise. The proposed method was evaluated on an artificial data and benchmark data. And we applied to the real-world response modeling for the insurance promotion provided by the CoIL Challenge 2000. The experimental results present the proposed method enables response modeling with only a few customer data without concerning the technical difficulties, selecting the best value of the hyperparameter and mitigating the influence of noise. | - |
- Appears in Collections:
- Special Graduate Schools > Graduate School of Science and Technology > Department of Industrial Engineering > 3. Theses(Master)
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
-
- License
- STATISTICS
- Total Visit :3,748,756
- Total Download :1,818
- Today View :26,415
