| 摘要 |
A method of determining if there is a relationship between measurement objects in a measurement data set is disclosed. The measurement data set consists of a set of measurement objects each having a given number of numeric measurement values. Furthermore, a measurement subset has been identified in the measurement data set by means of a given clustering method. The measurement subset consists of a subset of the measurement objects. Next, a value of cluster property index is calculated for the measurement subset. Furthermore, multiple reference data sets are generated. Each reference data sets are generated. Each reference data set consists of a set of reference objects each having randomly chosen numeric reference values. The number of numeric reference values for each reference object is equal to the number of numeric measurement values of a measurement object. Then reference subsets are identified in each of the multiple reference data sets by means of the given clustering method. Furthermore, values of the cluster property index for at least a subset of the reference subsets are calculated. Next the value of the cluster property index for said measurement subset is compared with the distribution of the values of the cluster property index for the subset of the reference subsets. The cluster properties of the measurement subset is then assessed based on this comparison, and it is determined whether there is a relationship between the measurement objects in the measurement data set based on the assessment of the cluster properties of the measurement subset.
|
