Profiling Process
When you begin a data integration project, profiling is often the first step. You can create profiles to analyze the content, quality, and structure of data sources. As a part of the profiling process, you discover the metadata of data sources.
You use different profiles for different types of data analysis, such as a column profile. You uncover and document data quality issues. Complete the following tasks to perform profiling:
- 1. Find and analyze the content of data in the data sources. Includes datatypes, value frequency, pattern frequency, and data statistics, such as minimum value and maximum value.
- 2. Review profiling results.
- 3. Create reference data.
- 4. Drill down on profile results.
- 5. Document data issues.
- 6. Create and run rules.
- 7. Create scorecards to monitor data quality.