Filemaker 16 or 17 Data Matching

We need to match a table with 500.000+ URLs wich include product names to another table which have keywords we can use to find the best matches. This has to be done within filemaker and I would prefer with MBS plug-in functions since it has specials feature in this area like Regex. Or probabaly even better trough Filemaker which is calling a seperate ElasticSearch Installation trough API. Example: [login to view URL] = Apple Iphone 8 Grey If you want to use ElasticSearch you must write the API Acess from FM to a local istance of ElasticSearch. What I need is a solid Solution in Filemaker that gives me the best matches and maybe the score of the matches so we can choose by hand the last percentage to match the right URLS. Also we need a translation Table to translate characters which are not alowed in URLs. For Example the Keyword "Google+" would be in a non URL encoded URL like "Google Plus" or striped to simple "Google". We Need to macth those tables on a monthly Basis. We can provide a example file with about 50 Keywords and about 1000 URLs for each Keyword, which we beleive which come close to the Keyword. Please explain shortly which technique you would use and why do you think its the best. Right now I would prefere Elastic Search because we need an API acees for another project anyway.