基于确定性转换的IFTS预测

Prediction of IFTS Based on Deterministic Transition

模糊时间序列预测模型在对不确定数据集的模糊趋势描述和论域划分方面有局限性,对此文中规范了直觉模糊时间序列的定义,应用直觉模糊C均值聚类算法优化论域区间划分,通过加入回溯机制构建确定性转换直觉模糊规则库,在此基础上提出一种直觉模糊时间序列预测方法,较好地反映了不确定系统数据的特征分布,提高了复杂环境下时间序列的预测精度.通过典型实例验证了该方法的有效性和优越性.

To break the limitation in the description of the fuzzy trend of uncertain data sets and the partitioning intervals, the definition of intuitionistic fuzzy time series is regulated. A forecasting method of intuitionistic fuzzy time series is proposed, which optimize the domain-dividing interval with an intuitionistic fuzzy C-means clustering algorithm. Deterministic transition intuitionistic fuzzy rules are established by adding a back-tracking scheme. The proposed method can better reflect the characteristic distribution of the uncertain system and improve the prediction accuracy of time series in complicated environments. Validity and superiority of the method are checked with a classical instance.