DEV Community

Discussion on: Algorithm explained: Text similarity using a vector space model

Collapse
 
thormeier profile image
Pascal Thormeier

Hey, that highly depends on your current database structure and technology. If you're using a relational database, such as Oracle, Postgres or MySQL, I would suggest the following database model:

Two tables, one containing the text and one linking text IDs together with their similarities as an extra column

You can put the text and some auto-generated ID in the table texts and store their similarities in the table similarities.

The exact queries for this, again, depend on your database structure and technology.

Does this help?

Collapse
 
endluen profile image
Endluen

Thank you so much ❤