Clustering support system, method, and program

Abstract

A classification means 181 generates a plurality of clusters by classifying subject data through machine learning. A display means 182 displays subject data included in the generated clusters in a predetermined mode at a position on a graph corresponding to the subject data in accordance with the feature of the subject data. At that time, the display means 182 changes the mode for displaying the subject data on the graph in accordance with numerical information for each feature.

Claims

1. A classification method that generates multiple clusters by classifying target data using machine learning, The system includes a display means that displays the target data included in the generated cluster in a predetermined manner at a corresponding position on a graph according to the feature quantities of the target data, The display means changes its mode of display on the graph to simultaneously show the numerical information of two or more of the features for the target data, according to the numerical information of each feature. A clustering support system characterized by the following features.

2. The display method involves overlaying and mixing the colors indicated by the numerical information of each feature to display the result in a mixed color display. The clustering support system according to claim 1.

3. The display means changes the intensity or brightness of the pattern according to the numerical information. A clustering support system according to claim 1 or claim 2.

4. The display means recalculates statistics for the target data within a specified range on the displayed graph, and changes the manner in which the target data within the specified range is displayed on the graph according to the recalculated statistics. A clustering support system according to claim 1 or claim 2.

5. The display means changes the appearance of the features based on the correlation between the specified features and displays them accordingly. A clustering support system according to claim 1 or claim 2.

6. The display method reduces the dimensionality of the target data and displays it on a graph. A clustering support system according to claim 1 or claim 2.

7. A computer generates multiple clusters by classifying the target data using machine learning, The computer displays the target data included in the generated cluster in a predetermined manner at the corresponding position on the graph according to the feature quantities of the target data. When the computer displays the data, it changes its behavior to show numerical information for two or more of the features simultaneously on the graph, according to the numerical information for each feature. A clustering support method characterized by the following features.

8. On the computer, A classification process that generates multiple clusters by classifying the target data using machine learning, and The generated clusters are subjected to a display process that shows the target data in a predetermined manner at the corresponding positions on the graph, according to the feature quantities of the target data. In the display process described above, the display method is changed to show the numerical information of two or more of the features simultaneously for the target data, according to the numerical information of each feature, on the graph. A clustering support program for this purpose.

Images