计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (2): 253-258.DOI: 10.3778/j.issn.1002-8331.1701-0271

• 工程与应用 • 上一篇    下一篇

粗糙集和支持向量机的表具识别算法研究

唐  亮1,2,仲元昌2,沈甲甲2,马天智2   

  1. 1.西南计算机有限责任公司,重庆 400060
    2.重庆大学 通信工程学院,重庆 400044
  • 出版日期:2018-01-15 发布日期:2018-01-31

Digital recognition based on rough set and support vector machine optimized by improved quantum-behaved particle swarm optimization

TANG Liang1,2, ZHONG Yuanchang2, SHEN Jiajia2, MA Tianzhi2   

  1. 1.Southwest Computer CO., LTD., Chongqing 400060, China
    2.College of Communication Engineering, Chongqing University, Chongqing 400044, China
  • Online:2018-01-15 Published:2018-01-31

摘要: 针对量子粒子群算法具有陷入局部值缺点,提出了一种基于改进量子粒子群算法优化的粗糙集和支持向量机相结合的表具识别算法,引入人工蜂群算法和免疫算法,来提高算法搜索空间、收敛速度。首先通过改进量子粒子群算法优化的粗糙集对得到的特征向量进行属性约简,然后经过改进量子粒子群算法优化支持向量机参数。最后通过实验仿真表明,改进的算法能有效地减少决策属性的个数,提高了粗糙集属性约简能力,优化了支持向量机的参数,算法收敛速度快,识别准确率高。

关键词: 表具识别, 量子粒子群算法, 粗糙集, 属性约简, 支持向量机

Abstract: In order to restrain particles from trapping in local optimum, this paper presents a new algorithm, which is based on the combination of rough sets and Support Vector Machine(SVM) optimized by Improved Quantum-behaved Particle Swarm Optimization(IQPSO), the artificial bee colony algorithm and immune algorithm are introduced to speed up the search space and improve the convergence speed. Firstly, the rough set optimized by IQPSO is used to reduce the attributes of the feature vector, and then the parameters of SVM are optimized by IQPSO. Finally, the simulation experiment shows that the improved algorithm can effectively reduce the number of decision attribute, improve rough set attribute reduction ability, obtain the better parameters of SVM, this algorithm has a fast convergence speed and high recognition accuracy.

Key words: gas meter recognition, quantum-behaved particle swarm optimization, rough set, attribute reduction, support vector machine