collect-initial-data

To document initial data collection for a CRISP-DM project, you generate an acquisition log, initial data inventory, and storage plans under docs/crisp-dm/2-data-understanding/ and notebooks/ to ensure reproducibility. This provides structured governance artifacts for onboarding datasets.

A data inventory is a structured record of received datasets, selection rationales, and loading plans generated during CRISP-DM phase 2. You need it to enable reproducible data science workflows, guide data onboarding, and establish governance for downstream analysis.

You create a data acquisition log and loading plan by profiling available datasets and documenting selection rationales. The process outputs notebook and docs artifacts that capture how data is acquired and stored, ensuring the data onboarding process is reproducible.

Yes, generating a data collection report requires reading existing CRISP-DM docs from phases 1.1 through 1.3. This prerequisite ensures the acquisition log and data inventory align with established business objectives and project context.

The best way to profile datasets and record selection rationale is to process available datasets through a structured CRISP-DM workflow. This generates a data collection report containing acquisition logs and inventory artifacts that guide governance and reproducibility.

Yes, you can inventory data across different data science projects because the approach applies generically to CRISP-DM phase 2. It produces standardized acquisition logs and storage plans that support reproducible workflows regardless of the specific domain.