現實生活中，往往因為樣本取得困難、過高的成本以及實際發生次數過少等情況，造成我們必須迫於面對少量的資料來做未來進一步的預測。因此，若能從小樣本中挖掘出更多重要的資訊，以提升其預測能力，是現今存在重要的研究議題。另一方面，由於小樣本資料的數量已經過少，每一筆資料的取得都不易，即便資料中存在著離群值，但也可能隱含重要的涵義。故本研究希望在不刪除任何一筆資料的情況下，藉由給予資料權重來重新定義每筆資料的重要程度。本研究方法有兩個部分，首先利用階層式分群法中的完全連結法對原始資料進行分群，透過資料內部的結構來找出重要特徵。第二個步驟，是將分群後產生的群集，用來建立權重式整體趨勢擴展函數，並計算各原始資料所對應的隸屬值，即產生新的屬性。透過原始資料與新產生的屬性結合成全新的資料集。接著利用一般普遍常使用的統計迴歸、倒傳遞類神經網路、支援向量迴歸等三個預測模型做驗證，以比較原始資料集與新資料集之間的預測能力。本研究選用兩個個案進行實驗例證，結果顯示本研究的方法不論從總誤差值、均方誤差等來觀察，都能有效降低，也代表能成功地有效提升小樣本的預測能力。

In real life, we are often forced to use a quite small amount of information to do further predictions, because of the difficulty to obtain samples from real cases. Therefore, how to find out more important information in a small data set to enhance its forecast ability is an important research issue. Besides, we know the acquisition of each data in small data set is not easy, even there exist a few outliers, thus in this study, we give a new weight to each data to define the importance of it. The proposed methods consists of two parts, the first part is using complete-linkage in hierarchical clustering techniques to separate the original data into clusters. The second parts is to build up the weighted-based data transformations function, in which new attributes are computed using fuzzy membership functions obtained by the corresponding membership grades in each cluster. Two real cases are selected to compare the proposed forecasting model with linear regression, back-propagation neural network, and support vector machine for regression methods. The result shows that the proposed method has better performance than using the raw data with regard to the total error and mean square error.

Amari, S., & Wu, S. (1999). Improving support vector machine classifiers by modifying kernel functions. Neural Networks, 12(6), 783-789.
Bresfelean, Vasile Paul. (2007). Analysis and predictions on students' behavior using decision trees in Weka environment. Paper presented at the Information Technology Interfaces, 2007. ITI 2007. 29th International Conference on.
Donetti, L., & Munoz, M.A. (2004). Detecting network communities: a new systematic and efficient algorithm. Journal of Statistical Mechanics: Theory and Experiment, 2004(10), P10012.
Gibbons, F.D., & Roth, F.P. (2002). Judging the quality of gene expression-based clustering methods using gene annotation. Genome research, 12(10), 1574-1581.
Hasegawa, T., Sekine, S., & Grishman, R. (2004). Discovering relations among named entities from large corpora. Paper presented at the Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics.
Huang, C., & Moraga, C. (2004). A diffusion-neural-network for learning from small samples. International Journal of Approximate Reasoning, 35(2), 137-161.
Huang, Zheng, Lv, Cong-wei, Yang, Wen-ying, & Zhang, Jian-huan. (2009). Nonlinear modeling of piezoelectric actuator based on SVR. Paper presented at the Piezoelectricity, Acoustic Waves, and Device Applications (SPAWDA) and 2009 China Symposium on Frequency Control Technology, Joint Conference of the 2009 Symposium on.
Johnson, S.C. (1967). Hierarchical clustering schemes. Psychometrika, 32(3), 241-254.
Karypis, G., Han, E.H., & Kumar, V. (1999). Chameleon: Hierarchical clustering using dynamic modeling. Computer, 32(8), 68-75.
Khan, NM, Ksantini, R., Ahmad, IS, & Boufama, B. (2012). A novel SVM+ NDA model for classification with an application to face recognition. Pattern Recognition, 45(1), 66-79.
Kuiper, F.K., & Fisher, L. (1975). 391: A Monte Carlo comparison of six clustering procedures. Biometrics, 777-783.
Li, D.C., Chang, C.C., & Liu, C.W. (2011a). Using structure-based data transformation method to improve prediction accuracies for small data sets. Decision Support Systems.
Li, D.C., Chang, C.J., Chen, C.C., & Chen, W.C. (2011b). Forecasting short-term electricity consumption using the adaptive grey-based approach–an Asian case. Omega.
Li, D.C., Chen, L.S., & Lin, Y.S. (2003). Using functional virtual population as assistance to learn scheduling knowledge in dynamic manufacturing environments. International Journal of Production Research, 41(17), 4011-4024.
Li, D.C., Fang, Y.H., Liu, C.W., & Juang, C. (2012a). Using past manufacturing experience to assist building the yield forecast model for new manufacturing processes. Journal of Intelligent Manufacturing, 23(3), 857.
Li, D.C., & Lin, Y.S. (2006). Using virtual sample generation to build up management knowledge in the early manufacturing stages. European Journal of Operational Research, 175(1), 413-434.
Li, D.C., & Liu, C.W. (2012). Extending Attribute Information for Small Data Set Classification. Knowledge and Data Engineering, IEEE Transactions on, 24(3), 452-464.
Li, D.C., Liu, C.W., & Hu, S.C. (2010). A learning method for the class imbalance problem with medical data sets. Computers in Biology and Medicine, 40(5), 509-518.
Li, D.C., Wu, C.S., Tsai, T.I., & Lina, Y.S. (2007). Using mega-trend-diffusion and artificial samples in small data set learning for early flexible manufacturing system scheduling knowledge. Computers & Operations Research, 34(4), 966-982.
Li, Der-Chiang, Chen, Wen-Chih, Liu, Chiao-Wen, & Lin, Yao-San. (2012b). A non-linear quality improvement model using SVR for manufacturing TFT-LCDs. Journal of Intelligent Manufacturing, 23(3), 835.
Mallios, Nikolaos, Papageorgiou, Elpiniki, & Samarinas, Michael. (2011). Comparison of Machine Learning Techniques using the WEKA Environment for Prostate Cancer Therapy Plan. Paper presented at the Enabling Technologies: Infrastructure for Collaborative Enterprises (WETICE), 2011 20th IEEE International Workshops on.
McCulloch, W.S., & Pitts, W. (1943). A logical calculus of the ideas immanent in nervous activity. Bulletin of mathematical biology, 5(4), 115-133.
Neter, J., Wasserman, W., & Kutner, M.H. (1996). Applied linear regression models (Vol. 3): Irwin Chicago.
Niyogi, P., Girosi, F., & Poggio, T. (1998). Incorporating prior information in machine learning by creating virtual examples. Proceedings of the IEEE, 86(11), 2196-2209.
Olson, Clark F. (1995). Parallel algorithms for hierarchical clustering. Parallel computing, 21(8), 1313-1325.
Rumelhart, D.E., Hinton, G.E., & Williams, R.J. (2002). Learning representations by back-propagating errors. Cognitive modeling, 1, 213.
Sanchez, A., & David, V. (2003). Advanced support vector machines and kernel methods. Neurocomputing, 55(1), 5-20.
Sneath, PHA. (1969). Evaluation of clustering methods. Numerical taxonomy, 257-270.
Tsai, Tung-I, & Li, Der-Chiang. (2008). Utilize bootstrap in small data set learning for pilot run modeling of manufacturing systems. Expert Systems with Applications, 35(3), 1293-1300.
Vapnik, V. (1999). The nature of statistical learning theory: springer.
Wang, H.F., & Huang, C.J. (2009). Data Construction Method for the analysis of the spatial distribution of disastrous earthquakes in Taiwan. International Transactions in Operational Research, 16(2), 189-212.
Yang, Haiyan, Zhou, Yongquan, & Liu, Hongxia. (2010). Chaos optimization svr algorithm with application in prediction of regional logistics demand. Advances in Swarm Intelligence, 58-64.
Yang, T., & Kecman, V. (2009). Adaptive local hyperplane algorithm for learning small medical data sets. Expert Systems, 26(4), 355-359.