A clustering based forecasting algorithm for multivariable fuzzy time series using linear combinations of independent variables

Dear Researcher, Thank you for using this code and datasets. I explain how CFTS code related to my paper "A clustering based forecasting algorithm for multivariable fuzzy time series using linear combinations of independent variables" published in Applied Soft Computing works. All datasets mentioned in the paper accompanied with CFTS code are included. If there is any question feel free to contact me at: [e-mail] [e-mail] Regards, S. Askari Guidelines for CFTS algorithm: 1. Open the file CFTS Code using MATLAB. 2. Enter or paste name of the dataset you wish to simulate in line 5 after "load". It loads the dataset in the workplace. 3. Lines 6 and 7: "r" is number of independent variables and "N" is number of data vectors used for training. 4. Line 9: "C" is number of clusters. You can use the optimal number of clusters given in Table 6 of paper or your own preferred value. 5. If line 28 is "comment", covariance norm (Mahalanobis distance) is use and if it is "uncomment", identity norm (Euclidean distance) is used. 6. Please press Ctrl Enter to run the code. 7. For your own dataset, please arrange the data as the datasets described in MS Word file "Read Me".尊敬的研究人员： 感谢您使用本代码与数据集。下文将说明与我发表于《应用软计算》（Applied Soft Computing）的论文《基于独立变量线性组合的多变量模糊时间序列聚类预测算法》相关的CFTS代码使用方法。本文提及的所有配套CFTS代码的数据集均已包含于此。如有任何疑问，可通过以下邮箱联系我：[e-mail]、[e-mail]。 顺颂时祺 S. Askari CFTS算法使用指南： 1. 使用MATLAB打开CFTS代码文件。 2. 在第5行"load"后输入或粘贴您拟仿真的数据集名称，即可将数据集加载至MATLAB工作区。 3. 第6、7行中，"r"代表独立变量的数量，"N"代表用于训练的数据向量总数。 4. 第9行中，"C"代表聚类簇数。您可使用论文表6中给出的最优聚类簇数，或自行设定偏好值。 5. 若第28行代码处于注释状态，则使用协方差范数（马氏距离，Mahalanobis distance）；若处于取消注释状态，则使用单位矩阵范数（欧氏距离，Euclidean distance）。 6. 按下Ctrl+Enter组合键即可运行代码。 7. 若使用自定义数据集，请按照MS Word文档"Read Me"中描述的数据集格式整理数据。