Machine learning (ML) models are an amalgamation of programming code with data, and they form the central concept in machine learning as they are what is being learned from the data to solve a given task. In addition to ML models, a functioning AI product contains other ingredients that make up a Machine Learning System, including tasks, features, and operations. The quality of an ML model is measured by its accuracy on test data using cross-validation.

In the context of materials science, ML models are used to build predictive models for various applications, such as predicting the melting temperatures of binary inorganic compounds. Building an ML model for materials properties prediction can be posed as a regression problem, where the goal is to predict continuous valued property values from a set of material attributes/features.

The ingredients or materials of an ML model depend on the specific problem domain, project time and other resources limitations, data quality, label quality and availability, dataset size, data domain variability, application infrastructure and architecture, and structure of the company. In the case of Kewpie Corporation, they used unsupervised machine learning to detect defective ingredients in their food products.