search

Schwittleymani / ECO

Electronic Chaos Oracle
https://schwittlick.net/eco
Apache License 2.0
6 stars 1 forks source link

Research some basic chatbot behaviour #152

Open schwittlick opened 7 years ago

schwittlick commented 7 years ago

a little bit of background and basics

https://apps.worldwritable.com/tutorials/chatbot/

machine learning chatbot implementation that should be checked regarding it can get additional training corpae

https://github.com/gunthercox/ChatterBot
schwittlick commented 7 years ago

a little excerpt of what the chatterbot is doing. its under active development but i am not really sure what it's doing

>>> chatbot.train("chatterbot.corpus.english")
>>> chatbot.get_response("Hello, how are you today?")
<Statement text:Fine, and you?>
>>> chatbot.get_response("Hello, how are you today?")
<Statement text:Fine, and you?>
>>> chatbot.get_response("Hello, how are you today?")
<Statement text:Fine, and you?>
>>> chatbot.get_response("Hello, how are you today?")
<Statement text:Fine, and you?>
>>> chatbot.get_response("What is going on?")                                                                                                 
<Statement text:I am baking a cake.>
>>> chatbot.get_response("That's funny.")                                                                                                     
<Statement text:The sky's up but I'm fine thanks. What about you?>
>>> chatbot.get_response("That's funny.")
<Statement text:Not much, how about you?>
>>> chatbot.get_response("That's funny.")
<Statement text:Not much, how about you?>
>>> chatbot.get_response("That's funny.")
<Statement text:The sky's up but I'm fine thanks. What about you?>
>>> chatbot.get_response("Loool")                                                                                                             
<Statement text:Greetings!>
>>> chatbot.get_response("Loool")
<Statement text:Greetings!>
>>> chatbot.get_response("Loool")
<Statement text:Greetings!>
>>> chatbot.get_response("Fuck off")                                                                                                          
<Statement text:Beautiful is better than ugly.>
>>> chatbot.get_response("Fuck off")
<Statement text:Beautiful is better than ugly.>
>>> chatbot.get_response("Fuck off")
<Statement text:Beautiful is better than ugly.>
>>> chatbot.get_response("Fuck away")                                                                                                         
<Statement text:No problem>
>>> chatbot.get_response("Fuck away")
<Statement text:No problem>
>>> chatbot.get_response("Fuck away")
<Statement text:No problem>
>>> chatbot.get_response("Fuck")
<Statement text:Hello, how are you today?>
>>> chatbot.get_response("Fuck")
<Statement text:Hello, how are you today?>
>>> chatbot.get_response("Fuck")
<Statement text:Hello, how are you today?>
>>> chatbot.get_response("Fuck yeah")
<Statement text:So what's your favorite color?>
>>> chatbot.get_response("Not gonna tell you")                                                                                                
<Statement text:Who? Who is but a form following the function of what>
>>> chatbot.get_response("What the fuck are you talking about")                                                               
<Statement text:I am baking a cake.>
>>> chatbot.train("chatterbot.corpus.english.conversations")
>>> chatbot.get_response("What the fuck are you talking about")
<Statement text:I am baking a cake.>

Downloading the UbuntuCorpusTrainer now, some conversation corpus from ubuntu or so, TBC

schwittlick commented 7 years ago

check out megahal implementations: https://github.com/pteichman/cobe