Machine Learning: Chapter 2

Concepts of machine learning

CRISP DM

is a data mining framework that provides a structured approach to planning, executing, and monitoring data mining projects.

Business Understanding

Determining business objectives, Assessing situation, determining data mining goals, produce project plan

Formulating the Problem

Determining the algorithm to be used

Determining the type of problem

Choosing which data mining tasks to perform, such as classification, regression, clustering, or anomaly detection.

Data Preparation

Collecting, Describing, Exploring and Verifying data

Modeling

Process is iterative and fluid

Iterations of feature engineering, training, evaluating and tuning

Evaluation

Evaluate results, review process, determine next steps

Deployment

Planning deployment, monitoring & maintenance, produce final report, review