Computer-readable storage medium storing data analysis program

A data analysis device, comprising:

• an input unit of inputting two time series cases of free correspondence comprising a plurality of data points each data point having one piece of time data and a plurality of attribute values, and an operation level indicating how each attribute works on a state of the time series case;
• and
• a data analysis unit of obtaining an optimum correspondence set between data points between the two time series cases by a processing system depending on the operation level, and obtaining a distance between the two time series cases using a processing system depending on the operation level, wherein
• the data analysis unit includes:
• a unit of calculating a distance between data points by calculating distances of substantially all combinations between data points between the two time series cases using the distance calculating system depending on an operation level, the unit of calculating a distance between data points outputting a first distance for a primary correspondence as a one-to-one correspondence, and a second distance for a secondary correspondence as a non-one-to-one correspondence when there is another primary correspondence for each attribute of each combination of data points;
• and
• a shortest distance calculating unit of calculating a shortest distance between the two time series cases using the distances of substantially all combinations of data points calculated by the unit of calculating a distance between data points, the shortest distance calculating unit obtaining the shortest distance between the two time series cases under a condition in which one or less of correspondences held by each data point is the primary correspondence, and another is the secondary correspondence.

A data analysis method, comprising:

• a receiving operation of receiving two time series cases of free correspondence comprising a plurality of data points, each data point having one piece of time data and a plurality of attribute values, and an operation level indicating how each attribute works on a state of the time series case, and
• a data analysis operation of obtaining an optimum correspondence set between data points between the two time series cases by a processing system depending on the operation level, and obtaining a distance between the two time series cases using the processing system depending on the operation level, wherein
• the data analysis operation includes:
• calculating a distance between data points by calculating distances of substantially all combinations between data points between the two time series cases using the distance calculating system depending on an operation level, the calculating of the distance between data points outputting a first distance for a primary correspondence as a one-to-one correspondence, and a second distance for a secondary correspondence as a non-one-to-one correspondence when there is another primary correspondence for each attribute of each combination of data points;
• and
• calculating a shortest distance between the two time series cases using the distances of substantially all combinations of data points calculated by the calculating of the distance between data points, the calculating of the shortest distance obtaining the shortest distance between the two time series cases under a condition in which one or less of correspondences held by each data point is the primary correspondence, and another is the secondary correspondence.