Application of Support Vector Machines to a Small-Sample Prediction

Minghui MENG, Chuanfeng ZHAO

Abstract


The support vector machines (SVMs) is one kind of novel small-sample machine learning methods based on solid theoretical background. Highly nonlinear regression and classification are their two applications. Different from conventional statistics methods, the SVMs employs the structural risk minimizing principle, which leads to high predication precision. For this method is not essentially related to probability measure and Law of Large Numbers, the final decision function is only determined by a small fraction of sample, called support vectors. Consequently, the complexity of computation only depends on the number of support vectors rather than the dimensions of the original sample space. In most occasions of oil and gas development, only small samples are available to predict the results of one measure. Introduction of SVMs into these applications can significantly improve prediction precision.


Keywords


Support vector machine; Statistical learning; Prediction; Kernel function

Full Text:

PDF

References


[1] Feng, Q. H., Chen, Y. W., & Guo, J. L. (2003). Studies on response prediction of conformance control treatment in water injectors. Petroleum Drilling and Production Technology, 25(6), 63-65.

[2] Chen, Y. Y., Yu, X. D., & Gao, X. H. (2004). An new method to solve nonlinear classification and regression problems (I). Journal of Applied Meteorology, 15(3), 345-353.

[3] Vapnik, V. N. (1995). The nature of statistical learning theory. New York, USA: Springer Verlag.

[4] Zhai, L., Li, Z. M., & Zhang, X. (2007). An optimization method for conformance control treatment in polymer injection based on SVMs. Oil and Gas Reservoir Geology and Recovery, 14(4), 88-91.

[5] Li, D., Wang, H. L., & Du, Z. X. (2006). Regression prediction with SVMs for urban civil water consumption. Journal of Tianjin University (Social Science Edition), 8(1), 64-66.

[6] Zhang, F., Zhang, X., & Zhang, L. (2007). Using SVMs to predict reservoir productivity. Journal of Southwest Petroleum University, 29(3), 24-27.

[7] Le, Y. X., & Yuan, Q. S. (2005). Application of SVMs to oil and gas reservoir parameter prediction. Natural Gas Industry, 25(12), 45-47.

[8] Mei, J. X., Duan, S., & Pan, J. B. (2002). Application of SVMs to small sample prediction. Journal of Wuhan University (Science Edition), 48(6), 732-735.




DOI: http://dx.doi.org/10.3968/7830

Refbacks

  • There are currently no refbacks.


Copyright (c) 2015 Advances in Petroleum Exploration and Development




Share us to:   


Reminder

  • How to do online submission to another Journal?
  • If you have already registered in Journal A, then how can you submit another article to Journal B? It takes two steps to make it happen:

1. Register yourself in Journal B as an Author

  • Find the journal you want to submit to in CATEGORIES, click on “VIEW JOURNAL”, “Online Submissions”, “GO TO LOGIN” and “Edit My Profile”. Check “Author” on the “Edit Profile” page, then “Save”.

2. Submission

  • Go to “User Home”, and click on “Author” under the name of Journal B. You may start a New Submission by clicking on “CLICK HERE”.

We only use three mailboxes as follows to deal with issues about paper acceptance, payment and submission of electronic versions of our journals to databases:
caooc@hotmail.com; aped@cscanada.net; aped@cscanada.org

 Articles published in Advances in Petroleum Exploration and Development are licensed under Creative Commons Attribution 4.0 (CC-BY)

ADVANCES IN PETROLEUM EXPLORATION AND DEVELOPMENT Editorial Office

Address:1020 Bouvier Street, Suite 400, Quebec City, Quebec, G2K 0K9, Canada.

Telephone: 1-514-558 6138
Website: Http://www.cscanada.net
Http://www.cscanada.org
E-mail:office@cscanada.net;  office@cscanada.org

Copyright © 2010 Canadian Research & Development Centre of Sciences and Cultures