METHOD FOR ESTIMATING REASON, METHOD FOR PREDICTION, METHOD FOR ESTIMATING ATTRIBUTE VALUE, REASON ESTIMATION DEVICE, PREDICTION DEVICE, ATTRIBUTE VALUE ESTIMATION DEVICE, AND PROGRAM
To provide a method for estimating a reason for an event that happens in an observation target on the basis of an attribute value of an observation target, a method for prediction, a method for estimating an attribute value, a reason estimation device, a prediction device, an attribute value estimat...
Gespeichert in:
1. Verfasser: | |
---|---|
Format: | Patent |
Sprache: | eng ; jpn |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | To provide a method for estimating a reason for an event that happens in an observation target on the basis of an attribute value of an observation target, a method for prediction, a method for estimating an attribute value, a reason estimation device, a prediction device, an attribute value estimation device, and a program.SOLUTION: The method for estimating a reason by an analyzer includes the steps of: setting a plurality of parameters for an observation target and acquiring observation data in which an attribute value showing an attribute of an observation target is set for each parameter or a temporary value is set for each parameter if the attribute is not known (Step 1); grouping the observation data on the basis of the combination of the parameters (Step S2); setting an evaluation item of an event (Step S3); setting an evaluation value for the evaluation item for the observation data (Step S4); calculating the representative value of the evaluation value of the observation data and the frequency on a group-by-group basis (Step S5); calculating a group statistic value formed by relating the values to each other (Step S6). The method determines the value of the parameter common to a group corresponding to an arbitrary group statistic value as a parameter that causes the event and the value of the parameter.SELECTED DRAWING: Figure 2
【課題】観測対象の属性値に基づいて、観測対象に生じる事象の要因推定方法、予測方法、属性値推定方法、要因推定装置、予測装置、属性値推定装置及びプログラムを提供する。【解決手段】解析装置による要因推定方法は、観測対象について複数のパラメータを設定し、各パラメータに前記観測対象の属性を示す属性値又は前記属性が不明な場合には仮の値を設定した観測データを取得しS1、パラメータの組み合わせに基づいて観測データをグルーピングしS2、事象の評価項目を設定しS3、観測データに対して評価項目についての評価値を設定しS4、グループごとに観測データの評価値の代表値と度数を算出しS5、これらを対応付けたグループ統計値を算出しS6、任意のグループ統計値に対応するグループに共通するパラメータの値を、事象の要因となるパラメータおよびその値として決定する。【選択図】図2 |
---|