Artificial intelligence model system and method for property prediction by applying feature engineering algorithm of material composition-process

摘要

The present invention relates to an apparatus and method for generating an artificial intelligence model for material property prediction to which a characteristic selection algorithm for composition-process of a material is applied, and more particularly, in an artificial intelligence model generating system for data-based prediction, composition-related data, process-related The data input unit 100 receives a data set including data and data related to physical properties, analyzes the data set, extracts material data and content data from the composition related data, and the process related data and physical properties A basic separation unit 200 for separating related data, a classification unit 300 for applying classification numbering to the material data extracted from the basic separation unit 200 based on the received physical property information to be predicted, the classification Using the material data to which classification numbering is applied by the unit 300 and the process-related data and material property-related data input and separated by the data input unit 100, select properties for the input material and process-related data The data generation unit 400 for generating training data and verification data for generating an artificial intelligence model through the training and verification of the model by the training data and verification data generated by the data generation unit 400 and the data It relates to a material property prediction artificial intelligence model generating apparatus to which a characteristic selection algorithm for the composition-process of the material, characterized in that it comprises a model generating unit 500 for generating an artificial intelligence model for base prediction.