粒子群算法求解基于广义Choquet-积分的多元非线性回归模型

Using PSO Algorithm to Solve Nonlinear Multiregression Based on Generalized Choquet Integrals

多元回归模型已经成为当前数据挖掘中重要的方法之一,而求解回归模型的关键问题是如何确定回归系数和模糊测度.针对以往使用遗传算法确定回归系数和模糊测度时间复杂度高和收敛速度较慢的问题,使用一种高效的搜索算法——粒子群算法求解基于广义Choquet-积分的多元非线性回归模型,分别在人工数据和真实数据上进行实验,对粒子群算法和遗传算法进行比较.结果表明,用粒子群算法求解该模型不仅比遗传算法收敛速度快,而且还能搜索到比遗传算法更优的解.

Multiregression is one of the most common approaches used in data mining. The crux of using a nonlinear multiregression model is how to optimally determine the unknown regression parameters that include values of the fuzzy measures. Considering the high complexity and low convergence speed of gennetic algorithm in solving nonlinear multiregression,we use PSO （Particle Swarm Optimization） to solve nonlinear multiregression based on generalized Choquet integrals. Comparative experiments are conducted on both simulated and real data sets,and the results show that PSO can converge faster and obtain optimal solution than genetic algorithm for solving nonlinear multiregression.