What is CRISP in Data Mining?
CRISP-DM (Cross-Industry Standard Process for Data Mining) is a widely used
framework thathelps in planning and executing data mining projects in a
structured way.
It is not owned or created by any single organization. Instead, it is a
proven and practicalapproach used across industries to solve business
problems using data.
CRISP-DM acts like a step-by-step guide (roadmap) that helps teams move from
a businessproblem to a data-driven solution.
Why is CRISP-DM Important?
CRISP-DM helps by:
- Providing a clear structure for projects
- Saving time through best practices
- Improving accuracy and results
- Helping teams stay focused on business goals
It ensures that data mining efforts are aligned with real business needs.
Key Feature of CRISP-DM
- It is flexible – steps don’t always follow a strict order
- Teams can go back and repeat steps when needed
- It can be customized based on the project
Example
If a company wants to detect fraud (like money laundering), they may:
- Focus more on data exploration and visualization
- Instead of building complex models
CRISP-DM allows such flexibility.
Phases of CRISP-DM
CRISP-DM consists of 6 main phases:
1. Business Understanding
This is the most important step.
Here, you define:
What problem are you solving?
What does the business want to achieve?
Key Activities:
Set clear business objectives
Define success criteria
Create a project plan
Example:
Business goal: Reduce customer churn
Data goal: Predict which customers may leave
Also Consider:
Available resources (people, tools, data)
Risks and constraints
Cost vs benefit
2. Data Understanding
In this phase, you collect and explore the data.
Key Activities:
Collect data from different sources
Understand data structure and format
Explore patterns and relationships
Check data quality
Questions to Ask:
Is the data complete?
Are there errors or missing values?
Is the data useful for the problem?
3. Data Preparation
This phase prepares the data for analysis.
Key Activities:
Select relevant data
Clean the data (handle missing values, errors)
Create new features (derived data)
Combine multiple datasets
Example:
Creating a new column:
Total Purchase = Price × Quantity
4. Modelling
Here, you build machine learning or data mining models.
Key Activities:
Choose modelling techniques (e.g., decision trees, neural networks)
Split data into training and testing sets
Train the model
Tune parameters
Output:
One or more models ready for evaluation
5. Evaluation
In this phase, you check if the model meets business goals.
Key Activities:
Evaluate model performance (accuracy, etc.)
Compare multiple models
Check if results solve the business problem
Important:
A model may be technically correct but not useful for business.
6. Deployment
This is the final phase where the solution is used in real life.
Key Activities:
Deploy the model (e.g., dashboard, system integration)
Monitor performance
Maintain and update the model
Create final reports and presentations
Example:
A churn prediction model used by a company to retain customers
