JPWO2020208445A5 - - Google Patents

Claims (16)

A method for achieving enhanced diversity and learning of ensemble models by a processor in a computing environment, comprising:
generating one or more data sets by combining one or more clusters of data points of the minority class with selected data points of the majority class;
generating one or more ensemble models from said one or more data sets using a supervised machine learning operation; and generating an event using said one or more ensemble models. A method comprising predicting the 2. The method of claim 1, further comprising determining scores for test data points by combining one or more predictions from the one or more ensemble models. extracting one or more features from the plurality of data points;
2. The method of claim 1, further comprising classifying data points of said plurality of data points into said majority class or said minority class according to said one or more characteristics. clustering a plurality of minority class data points into K clusters forming said minority class; and clustering each of said K clusters of said minority class with a random number of said majority class. 2. The method of claim 1, further comprising: augmenting by the selected data points of . 2. The method of claim 1, further comprising detecting one or more subclasses of said minority class. 2. The method of claim 1, further comprising using the supervised machine learning operation to split the minority class into the one or more clusters of data points. 2. The method of claim 1, further comprising labeling the one or more clusters of data points of the minority class according to predetermined criteria. A system for providing enhanced diversity and learning of ensemble models in a computing environment, comprising:
one or more computers containing executable instructions which, when executed, cause the system to:
generating one or more data sets by combining one or more clusters of data points of the minority class with selected data points of the majority class;
generating one or more ensemble models from said one or more data sets using a supervised machine learning operation; and generating an event using said one or more ensemble models. A system that predicts . 9. The system of claim 8, wherein the executable instructions determine scores for test data points by combining one or more predictions from the one or more ensemble models. The executable instructions are
extracting one or more features from the plurality of data points;
classifying data points of the plurality of data points into the majority class or the minority class according to the one or more characteristics;
9. A system according to claim 8. The executable instructions are
clustering a plurality of minority class data points into K clusters forming the minority class;
augmenting each of the K clusters of the minority class with a random number of the selected data points of the majority class;
9. A system according to claim 8. 9. The system of claim 8, wherein said executable instructions detect one or more subclasses of said minority class. 9. The system of claim 8, wherein the executable instructions use the supervised machine learning operation to divide the minority class into the one or more clusters of data points. 9. The system of claim 8, wherein the executable instructions label the one or more clusters of data points of the minority class according to predetermined criteria. A computer program product for causing a computer to perform the steps of the method according to any one of claims 1 to 7. 16. A computer-readable storage medium recording the computer program according to claim 15.