| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To provide a time series data prediction device for reducing time and economic costs for processing, while maintaining calculation of highly accurate prediction value. <P>SOLUTION: The time series data prediction device includes: a multiresolution decomposition unit 102 for performing a multiresolution analysis for time series data in which numerical values are represented in time series and calculating analytical values in a plurality of analysis levels as resolution data; a feature level selection unit 105 for selecting an analysis level to be used for the prediction value calculation as a feature level from the plurality of analysis levels; a model learning machine input unit 201 having a plurality of input interfaces; a model input association unit 103 for performing processing to input the resolution data related to the feature level into a corresponding input interface; and a model learning unit 202 for performing arithmetic operation on the basis of the resolution data input from the plurality of input interfaces to calculate the prediction value. <P>COPYRIGHT: (C)2010,JPO&INPIT</p> |
