Utilizing artificial intelligence to generate and update a root cause analysis classification model

A device, comprising:

• one or more memories;
• and
• one or more processors, communicatively coupled to the one or more memories, to:
• train a classification model with defect classifier training data to generate a trained classification model for performing a root cause analysis;
• process the defect classifier training data, with the trained classification model, to identify information indicating classes for defects indicated in the defect classifier training data;
• process the defect classifier training data, with a count vector, a cosine scorer, and a linear regression model, to identify information indicating priorities for the defects;
• process the defect classifier training data, with the count vector and the linear regression model, to identify information indicating rework efforts for the defects;
• process the information indicating the priorities and the rework efforts for the defects, with a Pareto analysis model, to select a set of the classes for the defects;
• calculate defect scores for the set of the classes for the defects,
• wherein the defect scores are calculated based on a quantity of the defects in the set of the classes and the priorities associated with the defects in the set of the classes;
• select a particular class, from the set of the classes for the defects, based on the defect scores;
• process a historical data set associated with a portion of the particular class to identify a root cause corrective action recommendation,
• wherein implementation of the root cause corrective action recommendation is to remove a defect associated with the particular class;
• process information indicating the defect associated with the particular class, with the trained classification model, to generate a predicted root cause corrective action recommendation for the defect associated with the particular class,
• wherein implementation of the predicted root cause corrective action recommendation is to remove some or all of the defect associated with the particular class;
• determine an effectiveness score for the predicted root cause corrective action recommendation based on the root cause corrective action recommendation,
• wherein the effectiveness score provides a measure of an effectiveness of the predicted root cause corrective action recommendation in removing the defect associated with the particular class, as compared to the root cause corrective action recommendation;
• retrain the trained classification model, based on the measure of the effectiveness of the predicted root cause corrective action recommendation in removing the defect associated with the particular class, to generate a retrained classification model;
• and
• perform one or more actions based on the retrained classification model.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors of a device, cause the one or more processors to:
• train a classification model with defect classifier training data to generate a trained classification model for performing a root cause analysis,
• wherein the defect classifier training data includes information indicating one or more of:
• descriptions of defects,
• classes for the defects,
• priorities for the defects, or
• rework efforts for the defects;
• process the defect classifier training data, with a count vector and a linear regression model, to identify the information indicating the rework efforts for the defects;
• process the information indicating the priorities and the rework efforts for the defects, with a Pareto analysis model, to select a set of the classes for the defects;
• calculate defect scores for the set of the classes for the defects,
• wherein the defect scores are calculated based on a quantity of the defects in the set of the classes and the priorities associated with the defects in the set of the classes;
• select a particular class, from the set of the classes for the defects, based on the defect scores;
• process a historical data set associated with a portion of the particular class to identify a root cause corrective action recommendation,
• wherein implementation of the root cause corrective action recommendation is to remove a defect associated with the particular class;
• process information indicating the defect associated with the particular class, with the trained classification model, to generate a predicted root cause corrective action recommendation for the defect associated with the particular class,
• wherein implementation of the predicted root cause corrective action recommendation is to remove some or all of the defect associated with the particular class;
• process the predicted root cause corrective action recommendation and the root cause corrective action recommendation, with the linear regression model, to determine an effectiveness score for the predicted root cause corrective action recommendation,
• wherein the effectiveness score provides a measure of an effectiveness of the predicted root cause corrective action recommendation in removing the defect associated with the particular class, as compared to the root cause corrective action recommendation;
• retrain the trained classification model, based on the measure of the effectiveness of the predicted root cause corrective action recommendation in removing the defect associated with the particular class, to generate a retrained classification model;
• and
• perform one or more actions based on the retrained classification model.