| Peer-Reviewed

Car Insurance Risk Assessment with Data Mining for an Iranian Leading Insurance Company

Received: 27 April 2014     Accepted: 17 May 2014     Published: 30 May 2014
Views:       Downloads:
Abstract

Today’s competitive market leads industry to a serious fight. This fight has guided some companies to a sightless selling. Insurance companies lose lots of money each year because of not profitable and risky customers which are attracted blindly. Risky customers are one of the most important treats to insurance companies; therefore some of these companies adopt a credit scoring and risk assessment approach for identifying profitable and risky customers. One of the most preferable methods for risk assessment is data mining. In this article, authors would demonstrate a risk assessment problem in an Iranian leading insurance company. Car insurance customers of this company have been analyzed with six different data mining algorithms (C5, Classification and Regression Tree, Neural Networks, Logistic Regression, Bayesian Networks and Support Vector Machines) in two different approaches. One of these approaches is a direct approach in which the target field (risk) is predicted directly with data mining algorithms and then an ensemble model comprised from them. The other one is an indirect approach in which the target field would be divided in five fields, then five different ensemble models is comprised for each new target field. Afterwards the model with the highest confidence predicts the target fields for a test data record. At the end of this article the better results of indirect model would be shown.

Published in International Journal of Business and Economics Research (Volume 3, Issue 3)
DOI 10.11648/j.ijber.20140303.12
Page(s) 128-134
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2014. Published by Science Publishing Group

Keywords

Assessment, Insurance Industry, Car Insurance, Data Mining, Insurance Risk

References
[1] Credit Scoring. [Internet]. [cited 2012 Nov 25]. Available from: http://epic.org/privacy/creditscoring/.
[2] What is a credit score? [Internet]. [cited 2012 Nov 25]. Available from: http://www.bankrate.com/finance/credit-cards/what-is-a-credit-score.aspx.
[3] Marikkannu P, Shanmugapriya K. Classification of customer credit data for intelligent cre-dit scoring system using fuzzy set and MC2 — Domain driven approach. Electronics Computer Technology (ICECT). 2011 410-414.
[4] What’s in my FICO Score. [Internet]. [cited 2012 Dec 25]. Available from: http://www.myfico.com/crediteducation/whatsinyourscore.aspx.
[5] Koeniger W. Labor Income Risk and Car Insurance in the UK. The GENEVA Papers on Risk and Insurance Theory. 2004;29(1):55-74.
[6] Gourieroux C. The Econometrics of Risk Classification in Insurance. The GENEVA Papers on Risk and Insurance Theory. 1999;24:119-137.
[7] Richaudeau D. Automobile Insurance Contracts and Risk of Accident: An Empirical Test Using French Individual Data. The GENEVA Papers on Risk and Insurance Theory. 1999;24(1):97-114.
[8] Han J, Kamber M, Pei J. Data Mining Concepts and Techniques. Waltham: Morgan Kaufmann; 2012.
[9] Li F, Xu J, Dou ZT, Huang YL. Data mining-based credit evaluation for users of credit card. Machine Learning and Cybernetics. 2004 2586- 2591.
[10] Yin Q, Lu K. Data mining based reduction on credit evaluation index of bank personal customer. Future Information Technology and Management Engineering (FITME). 2010 570-573.
[11] Hsu CF, Hung HF. Classification Methods of Credit Rating - A Comparative Analysis on SVM, MDA and RST. Computational Intelligence and Software Engineering. 2009 1-4.
[12] Wah YB, Ibrahim IR. Using data mining predictive models to classify credit card applicants. Advanced Information Management and Service (IMS). 2010 394- 398.
[13] Hammer PL, Kogan A, Lejeune MA. A logi-cal analysis of banks’ financial strength ratings. Expert Systems with Applications. 2012 7808-7821.
[14] Capotorti A, Barbanera E. Credit scoring analysis using a fuzzy probabilistic rough set model. Computational Statistics and Data Analysis. 2012 981–994.
[15] Feki A, Ishak AB, Feki S. Feature selec-tion using Bayesian and multiclass Support Vector Ma-chines approaches: Application to bank risk prediction. Expert Systems with Applications. 2012 3087–3099.
[16] García V, Marqués AI, Sánchez JS. On the use of data filtering techniques for credit risk prediction with instance-based models. Expert Systems with Applications. 2012 13267–13276.
[17] Feng W, Zhao Y, Deng J. Application of SVM Based on Principal Component Analysis to Credit Risk Assessment in Com-mercial Banks. Intelligent Systems. 2009 49-52.
[18] Min Z. Credit Risk Assessment Based on Fuzzy SVM and Principal Component Analysis. Web In-formation Systems and Mining. 2009 125-127.
[19] Kim Kj, HyunchulAhn. A corporate credit rating model using multi-class support vector machines with anordinal pairwise partitioning approach. Computers & Operations Research. 2012 1800–1811.
[20] Zhang D, Hifi M, Chen Q, Ye W. A Hybrid Credit Scoring Model Based on Genetic Pro-gramming and Support Vector Machines. Natural Com-putation. 2008 8-12.
[21] Chen W, Xiang G, Liu Y, Wang K. Credit risk Evaluation by hybrid data mining technique. Systems Engineering Procedia. 2012 194 – 200.
[22] Chen W, Ma C, Ma L. Mining the customer credit using hybrid support vector machine technique. Expert Systems with Applications. 2009 7611–7616.
[23] Liu X, Zhu X. Study on the Evaluation System of Individual Credit Risk in commercial banks based on data mining. Communication Systems, Networks and Applications (ICCSNA). 2010 308-311.
[24] Chiu JY, Yan Y, Xuedong G, Chen RC. A New Method for Estimating Bank Credit Risk. Technolo-gies and Applications of Artificial Intelligence (TAAI). 2010 503- 507.
[25] Wang G, Hao J, Mab J, Jiang H. A comparative assessment of ensemble learning for credit scoring. Expert Systems with Applications. 2011 223-230.
[26] Wang G, Ma J. A hybrid ensemble ap-proach for enterprise credit risk assessment based on Support Vector Machine. Expert Systems with Applica-tions. 2012 5325–5331.
[27] Marqués AI, García V, Sánchez JS. Two-level classifier ensembles for credit risk assessment. Expert Systems with Applications. 2012 10916–10922.
[28] Wirth R, Hipp J. CRISP-DM: To-wards a Standard Process Model for Data Mining. In: 4th International Conference on the Practical Application of Knowledge Discovery and Data Mining (KDD '96); 2000; Manchester, UK. p. 29-39.
[29] Quinlan JR. C4.5: Programs for Machine Learning. Morgan Kaufmann Publishers; 1993.
[30] Quinlan JR. Improved use of continuous attributes in c4.5. Journal of Artificial Intelligence Research. 1996 77-90.
[31] Loh WY. Classification and regression trees. John Wiley & Sons. 2011 14-23.
[32] Timofeev R. Classification and Re-gression Trees (CART) Theory and Applications. Berlin: CASE - Center of Applied Statistics and Economics, Humboldt University; 2004.
[33] Ruggeri F, Faltin F, Kenett R. Bayesian Networks. Encyclopedia of Statistics in Quality & Reliability. 2007 1-6.
[34] Stergiou C, Siganos D. NEURAL NETWORKS. [Internet]. [cited 2012 Nov 27]. Available from: http://www.doc.ic.ac.uk/~nd/surprise_96/journal/vol4/cs11/report.html#What is a Neural Net-work.
[35] Logistic Regression. [Internet]. [cited 2012 Nov 27]. Available from: http://userwww.sfsu.edu/efc/classes/biol710/logistic/logisticreg.htm.
[36] Logistic Regression. [Internet]. [cited 2012 Nov 28]. Available from: http://www.dtreg.com/logistic.htm.
Cite This Article
  • APA Style

    Seyed Behnam Khakbaz, Nastaran Hajiheydari, Marziyeh Pourestarabadi. (2014). Car Insurance Risk Assessment with Data Mining for an Iranian Leading Insurance Company. International Journal of Business and Economics Research, 3(3), 128-134. https://doi.org/10.11648/j.ijber.20140303.12

    Copy | Download

    ACS Style

    Seyed Behnam Khakbaz; Nastaran Hajiheydari; Marziyeh Pourestarabadi. Car Insurance Risk Assessment with Data Mining for an Iranian Leading Insurance Company. Int. J. Bus. Econ. Res. 2014, 3(3), 128-134. doi: 10.11648/j.ijber.20140303.12

    Copy | Download

    AMA Style

    Seyed Behnam Khakbaz, Nastaran Hajiheydari, Marziyeh Pourestarabadi. Car Insurance Risk Assessment with Data Mining for an Iranian Leading Insurance Company. Int J Bus Econ Res. 2014;3(3):128-134. doi: 10.11648/j.ijber.20140303.12

    Copy | Download

  • @article{10.11648/j.ijber.20140303.12,
      author = {Seyed Behnam Khakbaz and Nastaran Hajiheydari and Marziyeh Pourestarabadi},
      title = {Car Insurance Risk Assessment with Data Mining for an Iranian Leading Insurance Company},
      journal = {International Journal of Business and Economics Research},
      volume = {3},
      number = {3},
      pages = {128-134},
      doi = {10.11648/j.ijber.20140303.12},
      url = {https://doi.org/10.11648/j.ijber.20140303.12},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijber.20140303.12},
      abstract = {Today’s competitive market leads industry to a serious fight. This fight has guided some companies to a sightless selling. Insurance companies lose lots of money each year because of not profitable and risky customers which are attracted blindly. Risky customers are one of the most important treats to insurance companies; therefore some of these companies adopt a credit scoring and risk assessment approach for identifying profitable and risky customers. One of the most preferable methods for risk assessment is data mining. In this article, authors would demonstrate a risk assessment problem in an Iranian leading insurance company. Car insurance customers of this company have been analyzed with six different data mining algorithms (C5, Classification and Regression Tree, Neural Networks, Logistic Regression, Bayesian Networks and Support Vector Machines) in two different approaches. One of these approaches is a direct approach in which the target field (risk) is predicted directly with data mining algorithms and then an ensemble model comprised from them. The other one is an indirect approach in which the target field would be divided in five fields, then five different ensemble models is comprised for each new target field. Afterwards the model with the highest confidence predicts the target fields for a test data record. At the end of this article the better results of indirect model would be shown.},
     year = {2014}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Car Insurance Risk Assessment with Data Mining for an Iranian Leading Insurance Company
    AU  - Seyed Behnam Khakbaz
    AU  - Nastaran Hajiheydari
    AU  - Marziyeh Pourestarabadi
    Y1  - 2014/05/30
    PY  - 2014
    N1  - https://doi.org/10.11648/j.ijber.20140303.12
    DO  - 10.11648/j.ijber.20140303.12
    T2  - International Journal of Business and Economics Research
    JF  - International Journal of Business and Economics Research
    JO  - International Journal of Business and Economics Research
    SP  - 128
    EP  - 134
    PB  - Science Publishing Group
    SN  - 2328-756X
    UR  - https://doi.org/10.11648/j.ijber.20140303.12
    AB  - Today’s competitive market leads industry to a serious fight. This fight has guided some companies to a sightless selling. Insurance companies lose lots of money each year because of not profitable and risky customers which are attracted blindly. Risky customers are one of the most important treats to insurance companies; therefore some of these companies adopt a credit scoring and risk assessment approach for identifying profitable and risky customers. One of the most preferable methods for risk assessment is data mining. In this article, authors would demonstrate a risk assessment problem in an Iranian leading insurance company. Car insurance customers of this company have been analyzed with six different data mining algorithms (C5, Classification and Regression Tree, Neural Networks, Logistic Regression, Bayesian Networks and Support Vector Machines) in two different approaches. One of these approaches is a direct approach in which the target field (risk) is predicted directly with data mining algorithms and then an ensemble model comprised from them. The other one is an indirect approach in which the target field would be divided in five fields, then five different ensemble models is comprised for each new target field. Afterwards the model with the highest confidence predicts the target fields for a test data record. At the end of this article the better results of indirect model would be shown.
    VL  - 3
    IS  - 3
    ER  - 

    Copy | Download

Author Information
  • Faculty of Management, University of Tehran, Tehran, Iran

  • Faculty of Management, University of Tehran, Tehran, Iran

  • Faculty of Management, University of Tehran, Tehran, Iran

  • Sections