Question

I would like to create a database for synonymous and antonymous, but im not sure how to make the relationships between tables for better performance and also which index should i create.

First i thought:

table name: synonymous
id (int) primary key
name (varchar) text index
syn_group_id (int)
ant_group_id (int)  

So all the synonymous will have the same syn_group_id but this is not possible as all the words dont share the same synonymous, same happens for antonymous. So im totally lost on how to do this.

Answer 1

I would image you could simply have a structure like this:

table: Words
id(int, autoindex,primary,key)
word(varchar, unique index)

table: Synonyms
lookup_id, related_id (both int related to words table id field with compound primary key indexing)

table: Antonyms (same structure as synonyms)

The synonyms and antonyms tables are simply just many-to-many join tables related to the words table.

So a sample lookup query could be:

SELECT w.word FROM words as w INNER JOIN synonyms as s on w.id = s.related_id WHERE s.lookup_id = [LOOKUP_ID];