Gunther Cox

Interesting, I'll mark this as a possible bug then. I'm not aware of any issues currently effecting how the BestMatch adapter selects responses, but I'd happy to work on a...

I'm really interested in implementing this. I've done some experiments with creating 'personas' by custom-training multiple bots. However, I think the option to have ChatterBot learn based on weighted training...

I believe there is actually two cases here, both of which I'd like to implement. 1. One bot is having many different conversations with single users (https://github.com/gunthercox/ChatterBot/issues/372). 2. The bot...

Hi @cesarandreslopez, currently ChatterBot only returns a single response. I can definitely see the value in being able to set how may close responses are returned. I'd be happy to...

I'm reopening this ticket, the changes that I mentioned were never made.

It might be a good idea to weight the advantages of threading vs multiprocessing for this task. Most computers nowadays are built with multi-core processors. A multiprocessor implementation of this...

Hi @talentoscope, I'd be happy to accept contributions for the MinHash algorithm. In the mean time, I'd recommend Levenshtein distance for faster comparisons than Jaccard similarity (https://chatterbot.readthedocs.io/en/stable/comparisons.html#chatterbot.comparisons.LevenshteinDistance). ChatterBot's performance with...

Something similar to lemmatization is being added in 1.0 (using partial matches to reduce the size of the search set). The current reason that all statements are compared is because...

@azarezade Clustering does sound like an effective strategy. Are there any criteria you would suggest for creating each of the dialog clusters?

I'll have to do a bit of reading. `DBSCAN` sounds very promising and it looks like sklearn has a number of clustering algorithms available.