CAJ | 학술논문

针对粒子群优化(PSO)算法在自适应神经模糊推理系统(ANFIS)中的集成应用，提出对学习神经模型参数、隶属度函数参数进行改进优化的算法。该算法可增强模糊系统的近似精度和可解释性，提高系统的性能，进而发现更好的分类优化规则。算法经4个标准数据库的数据测试，结果表现出更好的性能，获得更好的分类效果，同时降低了系统时间复杂度。
침대입자군우화(PSO)산법재자괄응신경모호추리계통(ANFIS)중적집성응용，제출대학습신경모형삼수、대속도함수삼수진행개진우화적산법。해산법가증강모호계통적근사정도화가해석성，제고계통적성능，진이발현경호적분류우화규칙。산법경4개표준수거고적수거측시，결과표현출경호적성능，획득경호적분류효과，동시강저료계통시간복잡도。
For the particle swarm optimization (PSO) algorithm on the adaptive neuro fuzzy inference system (ANFIS) integrated ap-plications, put forward an improved algorithm by optimizing the learning of neural model parameters, the parameters of membership functions. The algorithm can improve the approximation accuracy and interpretability of fuzzy systems and the performance of the sys-tem. The proposed method has been tested on four standard dataset from UCI machine learning, i.e. Iris Flower, Haberman’s Survival Data, Balloon and Thyroid dataset. The results have shown better classification using the proposed PSO-ANFIS and the time complexity has reduced accordingly.