Cross-Industry Standard Process for Data Mining (CRISP-DM) is a standardised methodology used by data mining experts, composed of the following phases:
Business Understanding
Finalising the exact specification of the problem together with methods that evaluate the achievement of the project goal.
Data Understanding
Conducting a preliminary analysis that describes and explores the data provided by the client.
Data Preparation
Defining aggregation/preprocessing operations to build a dataset that can be easily exploited during the upcoming phases.
Modelling
Designing and training different Artificial Intelligence solutions to solve a business problem.
Evaluation
Using traditional metrics to evaluate the obtained models, and consequently; select the model that is most likely to provide the best solution.
Implementation
Defining a methodological pipeline that allows the end customer to use the selected model even on previously unseen data.
