PROBLEM TO BE SOLVED: To accurately extract a non-linear structure included in time series data signals having irregular fluctuation with respect to a processing method of time series data, a processing program of time series data and a processing system of time series data.SOLUTION: A plurality of first featured points featuring the variation of time series data signals having irregular fluctuation are extracted from those time series data signals (A20), and second feature points are set in the neighborhood of the plurality of first feature points (A30). Also, the first state of fluctuation to be acquired from the first feature points and the second state of fluctuation to be acquired from the second feature points are analyzed (A40, A50). Furthermore, the template of a non-linear structure is set based on the first state of fluctuation (A60). The feature points related with either the first state or the second state having the non-linear structure closer to the template are selected (A70).COPYRIGHT: (C)2009,JPO&INPIT【課題】時系列データの処理方法,時系列データの処理プログラム及び時系列データの処理システムに関し、不規則なゆらぎを持つ時系列のデータ信号に内在する非線形構造を正確に抽出する。【解決手段】不規則なゆらぎを持つ時系列のデータ信号の中から、該データ信号の変動を特徴付ける複数の第一特徴点を抽出し(A20)、該第一特徴点の近傍に第二特徴点を設定する(A30)。また、該第一特徴点から得られるゆらぎの第一状態と、該第二特徴点から得られるゆらぎの第二状態とを解析する(A40,A50)。さらに、該ゆらぎの第一状態に基づいて非線形構造のテンプレートを設定する(A60)。該第一状態及び該第二状態のうち該テンプレートにより近い非線形構造を有する何れか一方に係る特徴点を選択する(A70)。【選択図】図6
