Clustering processing apparatus and clustering processing method

A non-transitory computer readable medium having recorded therein a program for causing a computer to function as a clustering processing apparatus that groups data including a plurality of samples into clusters based on feature vectors of the samples, the apparatus comprising:

• N clustering means for grouping, for corresponding one of N data blocks resulting from dividing the data, samples included in the data block into clusters within the data block, each clustering means sequentially taking each sample included in the corresponding one of the N data blocks as a target sample, grouping the target sample into one of the clusters within the data block, storing, in a first storage means, cluster information including identification information on each cluster into which the samples are grouped within the data block, and storing, in second storage means, sample assignment information indicating the cluster to which the target sample belongs;
• cluster information transferring means for selecting, for each of the data blocks, cluster information on a cluster to be integrated from the cluster information stored in the first storage means when a predetermined condition is met, and transferring the selected cluster information from the first storage means to third storage means;
• and
• updating means for integrating clusters selected based on the cluster information stored in the third storage means into an integrated cluster and updating the sample assignment information stored in the second storage means based on information of the integrated clusters.

A clustering processing method executed by a clustering processing apparatus for grouping data including a plurality of samples into clusters based on feature vectors of the samples, comprising:

• a processor of the clustering processing apparatus performing the steps of:
• a clustering step of grouping, for each of N data blocks resulting from dividing the data, samples included in the data block into clusters within the data block by corresponding one of N clustering processors, each clustering processor sequentially taking each sample included in the corresponding one of the N data blocks as a target sample, grouping the target sample into one of the clusters within the data block, storing, in a first storage unit, cluster information including identification information on each cluster into which the samples are grouped within the data block, and storing, in a second storage unit, sample assignment information indicating the cluster to which the target sample belongs;
• a cluster information transferring step of selecting, for each of the data blocks, cluster information on a cluster to be integrated from the cluster information stored in the first storage unit when a predetermined condition is met, and transferring the selected cluster information from the first storage unit to a third storage unit;
• and
• an updating step of integrating clusters selected based on the cluster information stored in the third storage unit into an integrated cluster and updating the sample assignment information stored in the second storage unit based on information of the integrated clusters.