Understanding the data science development process

Data science uses a highly iterative, collaborative development process to identify those variables and metrics that might be better predictors of performance. This development process supports testing, experimenting, failing, learning, unlearning, and retrying using a combination of advanced analytic algorithms, data transformation, and data enrichment techniques necessary to discover those variables and metrics that might be better predictors of performance.The data science development process is a non-linear framework of rapid exploration, discovery, learning, testing, failing, and learning again. The data science development process drives collaboration between data scientists and data engineers and business and operational subject-matter experts and domain experts to ideate, explore, and test those variables and metrics that might be better predictors of performance.Figure 3.4 highlights the highly non-linear, recursive approach...