As a simple start me (Theis) and August had a pretty interesting talk. We could use a set with seen blip names and then use a map lookup to generalize terms (e.g go and golang are the same thing and thus get simplified to go using the map) you then do a contains on the set with this simplified map value and if yes then you do some conflict correction (could be ignore all but the first instance to start with) later maybe do some LLM stuff
Please see issue #67 for this^
Another interesting solution for further precision could be the use of a KMP string algorithm
As a simple start me (Theis) and August had a pretty interesting talk. We could use a set with seen blip names and then use a map lookup to generalize terms (e.g go and golang are the same thing and thus get simplified to go using the map) you then do a contains on the set with this simplified map value and if yes then you do some conflict correction (could be ignore all but the first instance to start with) later maybe do some LLM stuff
Please see issue #67 for this^
Another interesting solution for further precision could be the use of a KMP string algorithm