Igloo Corporation announced that it has completed patent registration that improves the performance of machine learning prediction models modeled through data and algorithms.

Through the application of this patent, it is expected that the efficiency of feedback work, which is a core process of machine learning, can be increased.

Machine learning, a field of artificial intelligence, is a technology that allows machines to make predictions about another data by creating their own judgment criteria based on learning about data.

In order to utilize this well, labeling work is required to attach characteristics to various data and convert them into a form suitable for the machine learning prediction model.

In addition, among the prediction results presented by the machine learning prediction model, a feedback process that selects satisfactory/disappointing prediction results and re-learning them by assigning different labels should also be accompanied.

The patent acquired by igloo Corporation this time focuses on efficiently selecting security events subject to feedback.

The predictive event index for each attack type is calculated based on the uncertainty index that can check the reliability of the predicted event and the explainable AI index that can determine the type of attack event, and compared with the previously stored feedback event screening criteria.