Squirrels have fuzzy tails.
0 Members and 1 Guest are viewing this topic.
My dream is that the forum will be for hard questions (since only people are smart enough to answer), while the chatbot is for quick answers (since it's instant).Some ideas I have for it . . . it'll remember previous questions you asked and try to narrow down what you are truly asking. It'll in theory remember when a question was asked before, and therefore not repeat the same answer again. It'll be able to compare/match your question to what was already asked in the forum. It'll also be able to figure out when two people are talking about the same question.All the chatbots I'm aware of are entirely programmed to fool you it is a person - not actually do anything useful or be an 'expert reference resource.' Try asking any of the below chatbots how to build a robot, and you'll see how useless and ambiguous the replies are:http://www.jeeney.com/http://users.auth.gr/~dkaras/http://alice.pandorabots.com/http://188.8.131.52/chat/And I'm not sure if php is the best way to go, it's just the easiest way I know.And you are right, I think I should put more effort into forcing users to do a search before posting a thread. I remember you mentioning that before . . . I'll start looking into that now.
The question is how do you want to program it? Do you want to program it with lots of info to start, or do you want it to "learn" like a child? I don't know how public these algorithms are, but I imagine they are relatively basic. It would be nearly impossible to program it with 100,000's of Q's & A's. A mix of both may be the best option.
QuoteThe question is how do you want to program it? Do you want to program it with lots of info to start, or do you want it to "learn" like a child? I don't know how public these algorithms are, but I imagine they are relatively basic. It would be nearly impossible to program it with 100,000's of Q's & A's. A mix of both may be the best option.I have no problem with programming in the answers. The beginner will typically only ask one of say 300 questions. I got about ~100 already in there.There are currently 7,659 threads in this forum. Quite a few are repeat questions. Can't be that hard to extract the most frequent 10% It'll just take a programmer better than me, and with free time, to figure out how to match known questions to known answers intelligently. Now that I think about it, it's like using googles 'I'm feeling lucky' button, and actually getting a useful result on every try The domain is very specialized, so can't be too hard?
Technology Review is reporting on IBM's plans to take on Trebek at his own game. The 'Watson' computer system uses natural-language processing techniques to break down questions into their structural components and then search its database for relevant answers. A televised matchup with Trebek is planned for next year. 'David Ferrucci, the IBM computer scientist leading the effort, explains that the system breaks a question into pieces, searches its own databases for "related knowledge," and then finally makes connections to assemble a result. Watson is not designed to search the Web, and IBM's end goal is a system that it can sell to its corporate customers who need to make large quantities of information more accessible.'