高阶直觉模糊时间序列预测模型

High order intuitionistic fuzzy time series forecasting model

提出一种高阶直觉模糊时间序列预测模型。模型首先应用模糊聚类算法实现论域的非等分划分;然后,针对直觉模糊时间序列的数据特性,提出一种更具客观性的直觉模糊集隶属度和非隶属度函数的确定方法;最后,利用直觉模糊多维取式推理建立高阶模型的预测规则,进行预测。在Alabama大学入学人数和北京市日均气温2组数据集上分别与典型方法进行对比实验,结果表明该模型有效提高了预测精度,证明了模型的有效性和优越性。

A high order intuitionistic fuzzy time series forecasting model was built. In the new model, fuzzy clustering algorithm was used to get unequal intervals, and a more objective technique for ascertaining membership and non-membership functions of intuitionistic fuzzy set was proposed. On these bases, forecasting rules based on multi-dimension intuitionistic fuzzy modus ponens inference were established. At last, contrast experiments on the enrollments of the university of Alabama and the daily average temperature of Beijing were carried out, which show that the novel model has a clear advantage of improving the forecasting accuracy.