/schema-matching

Match schema attributes of relational databases by value similarity. As a study assignment, this isn't well documented, but you can contact me for questions and I may even add docs, if I sense enough interest.

Primary LanguagePythonMIT LicenseMIT

Schema Matching

Match schema attributes by value similarity.

Usage

Run ./schema-matching --help to see a usage description.

Examples

See the shell scripts in demo. I suggest that you start with 1-match.sh for something simple. The output will be, in this order,

  1. the norms between each column pair, i. e. a measure of how different they are, between 0 (identical) and 1 (completely different),

  2. the norm of the most closely matching column mapping, and

  3. the most closely matching column mapping, one pair per line.

1. and 2. require at least one level of verbosity (using -v or --verbose).

Pre-requisites

  • Python 3 (tested with v3.6.8)