skills.rest
lincolnadams5
lincolnadams5COMMUNITY
Explore all skills by this author →
GitHub

crisp-dm-pipeline

Automates construction of CRISP-DM machine learning pipelines with staged interviews.

Updated Apr 6, 2026
Search Tags:#data science#machine learning#evaluation#pipeline#modeling#CRISP-DM
One-click install
npx skills add https://github.com/lincolnadams5/intex-w26 --skill crisp-dm-pipeline-lincolnadams5
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: crisp-dm-pipeline
Source: https://github.com/lincolnadams5/intex-w26/tree/main/pipeline
Command: npx skills add https://github.com/lincolnadams5/intex-w26 --skill crisp-dm-pipeline-lincolnadams5

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, scikit-learn, xgboost, numpy, matplotlib, seaborn, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the construction of machine learning pipelines, ensuring adherence to the CRISP-DM (Cross-Industry Standard Process for Data Mining) framework, which guides the entire process from problem framing to evaluation.

Core Features & Use Cases

  • Staged Interview Protocol: Follows a structured approach to gather information and context for each phase of the CRISP-DM framework.
  • Modular Architecture: Utilizes a modular structure for reusable logic and independent notebooks, ensuring consistency and independence.
  • Independent Notebooks: Each phase notebook is self-contained and runnable independently, with a comprehensive Master notebook for stakeholder review.
  • Sign-off Gates: Ensures explicit user sign-off at each phase before proceeding to the next.
  • Problem Type Adaptation: Handles various machine learning problems including classification, regression, and time series forecasting with specific adaptations.
  • Imbalanced Data Handling: Provides strategies for handling imbalanced datasets, including class weights, SMOTE, and threshold tuning.
  • Evaluation Metrics: Utilizes appropriate evaluation metrics based on problem type and business objectives.

Quick Start

Run the skill by executing 'crisp-dm-pipeline' followed by your project's goal, e.g., 'crisp-dm-pipeline churn_prediction'.

Frequently Asked Questions about crisp-dm-pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a machine learning pipeline using the CRISP-DM framework?▼

To build a machine learning pipeline using the CRISP-DM framework, you can automate the construction process with structured staged interviews and modular architecture, ensuring consistent progression from problem framing to model evaluation.

What is the best way to handle imbalanced datasets in a scikit-learn pipeline?▼

Handling imbalanced datasets in a scikit-learn pipeline is achieved through specific strategies like applying class weights, utilizing SMOTE oversampling, and performing threshold tuning to optimize evaluation metrics.

Do I need pandas and xgboost installed to automate modeling pipelines?▼

Yes, you need pandas, xgboost, scikit-learn, numpy, matplotlib, and seaborn installed in your Python environment to support the framework's data processing, model training, and structured evaluation capabilities.

Can I run individual CRISP-DM phase notebooks independently for evaluation?▼

Yes, you can run each phase notebook independently because the modular architecture provides self-contained notebooks for isolated execution, alongside a comprehensive Master notebook for stakeholder review.

Does this framework support time series forecasting and regression problems?▼

Yes, the framework supports time series forecasting and regression problems by adapting its problem type handling, ensuring appropriate evaluation metrics are utilized based on specific business objectives.

Why does the modeling process require explicit sign-off gates between phases?▼

Explicit sign-off gates are required to ensure user validation before proceeding to the next phase, maintaining adherence to the CRISP-DM framework and securing structured alignment with business objectives.

Related Skills

Explore More Data & Analytics Skills→
omicverseomicverse
32

data-transform

Clean, reshape, normalize and engineer features from CSV/TSV/Parquet inputs using pandas and numpy locally.

Official
Advanced
zadnan2002zadnan2002

evaluate-model

Calculate model metrics and generate accuracy, calibration, and stability visualizations.

Community
Intermediate
jkitchinjkitchin
34

python-plotting

Select and apply Matplotlib, Seaborn, or Plotly for Python visualizations.

Community
Intermediate
thbraetthbraet

construct-data

Derive and transform features from cleaned datasets for CRISP-DM workflows.

Community
Advanced