The New Development in Support Vector Machine Algorithm Theory and Its Application

<span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US">As to classification problem, this paper puts forward the combinatorial optimization least squares support vector machine algorithm (COLS-SVM). Base</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US">d</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US"> on algorithmic analysis of COLS-SVM</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US"> and improves on it</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US">, </span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US">the improved COLS-SVM</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US"> can be used on individual credit evaluation. As to regression problem, appropriate kernel function and parameters were selected based on the analysis of support vector regression (SVR) algorithm. This paper proposes the forecasting model of coal mine ground-water-level based on SVR algorithm</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US"> and improves on it</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US">. In another regression problem, it </span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US">improves on </span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US">successive overrelaxation for support vector regression (SORR) algorithm to measure the cholesterol content of a blood sample concerning the three kinds of plasma lipoproteins (VLDL, LDL, HDL) in medical science.</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US"> </span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US">The numerical experiment results show that the </span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US">improved</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US"> COLS-SVM algorithm and Mine Ground-water-level Forecasting </span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US">improved</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US"> Model and </span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;" lang="EN-US">improved</span><span style="font-size: 10pt; font-family: "Times New Roman"; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 宋体; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US"> SORR algorithm are effective.</span>