預測模型之建模樣本的篩選方法及其電腦程式產品METHOD FOR SCREENING SAMPLES FOR BUILDING PREDICTION MODEL AND COMPUTER PROGRAM PRODUCT THEREOF

專利類型

發明

專利國別 (專利申請國家)

中華民國

專利申請案號

100147447

專利證號

I 451336

專利獲證名稱

預測模型之建模樣本的篩選方法及其電腦程式產品METHOD FOR SCREENING SAMPLES FOR BUILDING PREDICTION MODEL AND COMPUTER PROGRAM PRODUCT THEREOF

專利所屬機關 (申請機關)

國立成功大學

獲證日期

2014/09/01

技術說明

一種預測架構模型之建模樣本的篩選方法及其電腦程式產品。當一動態移動視窗(Dynamic Moving Window；DMW)中有加入一筆新樣本資料時，先對視窗內的所有樣本資料進行分群步驟，使得特性相近的樣本資料被歸類於同一群內。然後，檢視各個群集內之樣本資料的數目。若最大群集之樣本資料的數目大於預設門檻，則代表最大群集有許多相關樣本資料，故可摒棄此最大群集內之最舊樣本資料。若最大群集之樣本資料的數目小於或等於預設門檻，則代表最大群集中之樣本資料相當獨特，而必須被保留來建立或更新預測架構模型。 A method for screening samples for building a prediction scheme model and a computer program product thereof are provided. When a set of new sample data is added to a dynamic moving window (DMW), a clustering step is performed with respect to all of the sets of sample data within the window for grouping the sets of sample data with similar properties as one group. Then, the number of the the sets of sample data in each group is checked. If the number of the sets of sample data in the largest group is greater than a predetermined threshold, it means that there are many sets of sample data in the largest group, and the oldest sample data in the largest group can be deleted. If the number of the sets of sample data in the largest group is smaller than or equal to a predetermined threshold, it means that the sample data in the largest group are quite unique, and should be kept for building or refreshing the prediction scheme model.

備註

連絡單位 (專責單位/部門名稱)

企業關係與技轉中心

連絡電話

06-2360524

網址