基于自适应边界向量提取的多尺度v-支持向量机建模-控制与决策.pdfVIP

绗?0 鍗 绗? 鏈 鎺 鍒 涓 鍐 绛 2015 骞 4 鏈 Vol. 30 No. 4 Control and Decision Apr. 2015 鏂囩珷缂栧彿: 1001-0920 (2015) 04-0721-06 DOI: 10.13195/j.kzyjc.2014.0044 鍩轰簬鑷€傚簲杈圭晫鍚戦噺鎻愬彇鐨勫灏哄害饊仯-鏀寔鍚戦噺鏈哄缓妯 鑻忔垚鍒? 閮戝崥鍏? 鏉 骞 (杈藉畞鐭虫补鍖栧伐澶у淇℃伅涓庢帶鍒跺伐绋嬪闄紝杈藉畞鎶氶『113001) 鎽 瑕? 閽堝饊仯-鏀寔鍚戦噺鏈?饊仯-SVM) 鐢ㄤ簬澶ц妯°€佸宄版牱鏈缓妯℃椂鏄撳嚭鐜拌缁冮€熷害鎱㈠拰鍥炲綊绮惧害浣庣殑闂, 鎻愬嚭鍩 浜庤竟鐣屽悜閲忔彁鍙栫殑澶氬昂搴︷€仯-SVM 寤烘ā鏂规硶. 璇ユ柟娉曢噰鐢ㄤ竴绉嶈嚜閫傚簲杈圭晫鍚戦噺鎻愬彇绠楁硶, 浠庤缁冩牱鏈腑棰勬彁鍙栧嚭鍖呭惈 鍏ㄩ儴鏀寔鍚戦噺鐨勮竟鐣屽悜閲忛泦, 浠ョ缉鍑忚缁冩牱鏈妯? 骞堕€氳繃姹傝В澶氬昂搴︷€仯-SVM 浜屾瑙勫垝闂鑾峰彇鍏ㄥ眬鏈€浼樺洖褰掓ā鍨? 浠庡涓昂搴︿笂瀵瑰鏉傚垎甯冩牱鏈繘琛岄€艰繎. 浠跨湡缁撴灉琛ㄦ槑, 鍩轰簬杈圭晫鍚戦噺鎻愬彇鐨勫灏哄害饊仯-SVM 姣旔€仯-SVM 鍏锋湁鏇村ソ鐨 鍥炲綊缁撴灉. 鍏抽敭璇岤 澶ф牱鏈缓妯★紱杈圭晫鍚戦噺鎻愬彇锛涘灏哄害瀛︿範锛涴€仯-鏀寔鍚戦噺鏈 涓浘鍒嗙被鍙耳 TP181 鏂囩尞鏍囧織鐮? A Multiscale 饊仯-support vector machine modeling based on adaptive boundary vector extraction SU Cheng-li, ZHENG Bo-yuan, LI Ping (School of Information and Control Engineering锛孡iaoning Shihua University 锛孎ushun 113001 锛孋hina 锛嶤orrespondent: SU Cheng-li锛孍-mail 锛歴uchengli@lnpu.edu.cn) Abstract 锛氾細锛欰 multiscale 饊仯-support vector machine(饊仯-SVM) based on adaptive boundary vector extraction is presented. It overcomes the disadvantages of the slow training speed and the low regression accuracy which are caused by using the general 饊仯-SVM for large-scale and multi-peak sample modeling. An adaptive boundary vector extraction algorithm is used to extract the boundary vectors which include all support vectors from the training samples, so that reduces the sample scale. The global optimal regression model is obtained by solving the multiscale 饊仯-SVM quadratic programming problems, and the complex distribution sample can be approximated from multiple scales by the model. Simulation results