论文部分内容阅读
对国内外土地集约利用评价的相关文献研究,在支持向量机、蚁群算法基础上,提出相关系数、蚁群算法与支持向量机相结合评价方法,对指标进行相关分析,确定指标集,运用蚁群算法,优化支持向量机参数,得出较好的惩罚因子C,核函数σ和不敏感系数ε,再对支持向量机训练,该方法提高了训练准确度,对土地集约利用进行c ACO-SVM评价,并与ACO-SVM、GA-SVM的土地集约利用评价进行比较,评价与仿真结果表明,c ACO-SVM的土地集约利用评价优于ACO-SVM、GA-SVM两种方法,c ACO-SVM的土地集约利用评价效果比较理想。
On the basis of support vector machine and ant colony algorithm, this paper puts forward the evaluation method combining the correlation coefficient, ant colony algorithm and support vector machine to carry on the correlation analysis to the index, determine the index set, Ant colony algorithm, the parameters of support vector machine are optimized, the better penalty factor C, kernel function σ and insensitive coefficient ε are obtained, then the training of SVM is improved. The method improves the training accuracy, SVM, and compared with the evaluation of intensive land use of ACO-SVM and GA-SVM. The evaluation and simulation results show that the evaluation of intensive land use of c ACO-SVM is superior to that of ACO-SVM and GA-SVM The evaluation of intensive land use of ACO-SVM is quite satisfactory.