eprintid: 1329 rev_number: 12 eprint_status: archive userid: 2 dir: disk0/00/00/13/29 datestamp: 2015-02-20 12:00:01 lastmod: 2017-02-08 12:21:44 status_changed: 2015-04-27 12:10:42 type: article metadata_visibility: show creators_name: Zhou, Zhi-xiang creators_name: Han, Feng-qing corp_creators: College of Civil Engineering and Architecture, Chongqing Jiaotong University, Chongqing 400074, P. R. China corp_creators: School of Science, Chongqing Jiaotong University, Chongqing 400074, P. R. China title: An iterative modified kernel based on training data ispublished: pub subjects: SS divisions: EPOS-P full_text_status: none keywords: support vector regression, data-dependent, kernel function, iteration abstract: To improve performance of a support vector regression, a new method for a modified kernel function is proposed. In this method, information of all samples is included in the kernel function with conformal mapping. Thus the kernel function is data-dependent. With a random initial parameter, the kernel function is modified repeatedly until a satisfactory result is achieved. Compared with the conventional model, the improved approach does not need to select parameters of the kernel function. Simulation is carried out for the one-dimension continuous function and a case of strong earthquakes. The results show that the improved approach has better learning ability and forecasting precision than the traditional model. With the increase of the iteration number, the figure of merit decreases and converges. The speed of convergence depends on the parameters used in the algorithm. date: 2009 date_type: published publication: Applied Mathematics and Mechanics volume: 30 number: 1 publisher: Springer Verlag pagerange: 121-128 id_number: doi:10.1007/s10483-009-0113-x refereed: TRUE issn: 0253-4827 official_url: http://dx.doi.org/10.1007/s10483-009-0113-x access_IS-EPOS: limited owner: Publisher citation: Zhou, Zhi-xiang and Han, Feng-qing (2009) An iterative modified kernel based on training data. Applied Mathematics and Mechanics, 30 (1). pp. 121-128. DOI: https://doi.org/10.1007/s10483-009-0113-x