|
View:
New views
2 Messages
—
Rating Filter:
Alert me
|
|
|
assistance with an AIML ideaHi,
Im using python for my bots that use AIML and I was wondering about something. I usually keep logs of what my bot says and its responses from other people. Now what Im thinking is having another python script that goes through these responses and generate or update its current AIML markups based on reversing the roles of responses and the statements by and from the responder (does this make sense?) For example if someone asked if today is sunday and the bot's response is that it doesnt know and the responder tells it to ignore it. The script will update the AIML so that if someone tells the bot that they dont know what day it is today, the bot will tell them to ignore it. Has this been done before? Im just thinking in a very general manner here, im sure there are alot of ways in which this can be improved. Any comments or thoughts or obstacles with this aside from the obvious overfitting/overspecialized AIML; but I intend to counter this with more general chats. Cheers |
|
|
Re: assistance with an AIML ideaReusing the user's responses as future bot responses is a concept that has
come up a number of times. The mailing list archives will contain a number of references. However, the core of the debate mostly surrounds around the idea whether it is a good idea to have regular users update or add to your knowledgebase. Those who have experimented will have mixed experiences. Of course, there will be some real gems of answers added but it will mostly be of a different tone then what has been written already and often downright insulting. If you go this route then thread carefully. Kind regards, Anne. -----Oorspronkelijk bericht----- Van: alicebot-general-bounces@... [mailto:alicebot-general-bounces@...] Namens traltixx Verzonden: zaterdag 16 februari 2008 12:19 Aan: alicebot-general@... Onderwerp: [alicebot-general] assistance with an AIML idea Hi, Im using python for my bots that use AIML and I was wondering about something. I usually keep logs of what my bot says and its responses from other people. Now what Im thinking is having another python script that goes through these responses and generate or update its current AIML markups based on reversing the roles of responses and the statements by and from the responder (does this make sense?) For example if someone asked if today is sunday and the bot's response is that it doesnt know and the responder tells it to ignore it. The script will update the AIML so that if someone tells the bot that they dont know what day it is today, the bot will tell them to ignore it. Has this been done before? Im just thinking in a very general manner here, im sure there are alot of ways in which this can be improved. Any comments or thoughts or obstacles with this aside from the obvious overfitting/overspecialized AIML; but I intend to counter this with more general chats. Cheers -- View this message in context: http://www.nabble.com/assistance-with-an-AIML-idea-tp15517177p15517177.html Sent from the Alicebot General mailing list archive at Nabble.com. _______________________________________________ This is the alicebot-general mailing list Reply to alicebot-general@... Unsubscribe and change preferences at http://list.alicebot.org/mailman/listinfo/alicebot-general Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html Learn to read at http://www.literacy.org/ _______________________________________________ This is the alicebot-general mailing list Reply to alicebot-general@... Unsubscribe and change preferences at http://list.alicebot.org/mailman/listinfo/alicebot-general Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html Learn to read at http://www.literacy.org/ |
| Free Forum Powered by Nabble | Forum Help |