Method for training and using a classification model with association rule models

A method for applying a classification model on input data, wherein the input data contains a plurality of items associated with an entity, the method comprising:

• checking, using a processor of a computer, a set of rules from the classification model to determine which rules are applicable for the entity;
• aggregating lift values of the applicable rules from the classification model;
• and
• predicting a classification for the entity based on the aggregated lift value and the classification model.

A computer program product for training a classification model for detecting patterns in input data, wherein the computer program product comprises:

• a computer readable storage medium having computer readable program code embodied thereon, the computer readable program code, when executed by a processor of a computer, is configured to perform:
• retrieving a set of input data containing a plurality of items associated with a plurality of entities;
• adding a classification to each entity of the plurality of entities;
• computing rules from the set of input data and each classification, wherein each rule associates the classification of an entity with items associated with the entity;
• for each entity,
• determining which rules are applicable;
• and
• aggregating lift values of the applicable rules;
• and
• using the aggregated lift values, the plurality of entities, and each classification to determine a classification model.

A computer program product for applying a classification model on input data, wherein the input data contains a plurality of items associated with an entity, wherein the computer program product comprises:

• a computer readable storage medium having computer readable program code embodied thereon, the computer readable program code, when executed by a processor of a computer, is configured to perform:
• checking a set of rules from the classification model to determine which rules are applicable for the entity;
• aggregating lift values of the applicable rules from the classification model;
• and
• predicting a classification for the entity based on the aggregated lift value and the classification model.

A computer system for training a classification model for detecting patterns in input data, wherein the computer system comprises:

• a processor;
• and
• storage for storing computer readable program code, and wherein the processor is configured to execute the computer readable program code to perform:
• retrieving a set of input data containing a plurality of items associated with a plurality of entities;
• adding a classification to each entity of the plurality of entities;
• computing rules from the set of input data and each classification, wherein each rule associates the classification of an entity with items associated with the entity;
• for each entity,
• determining which rules are applicable;
• and
• aggregating lift values of the applicable rules;
• and
• using the aggregated lift values, the plurality of entities, and each classification to determine a classification model.

A computer system for applying a classification model on input data, wherein the input data contains a plurality of items associated with an entity, wherein the computer system comprises:

• a processor;
• and
• storage for storing computer readable program code, and wherein the processor is configured to execute the computer readable program code to perform:
• checking a set of rules from the classification model to determine which rules are applicable for the entity;
• aggregating lift values of the applicable rules from the classification model;
• and
• predicting a classification for the entity based on the aggregated lift value and the classification model.