On 07/21/2017 04:58 PM, [email protected] wrote:
Ah thanks for letting me know about those rules, one more thing about this tho. Is there already a rule which let's opencog(chatbot) know that any statement made by a person or other information source it is talking to is what they think instead of basically storing it in the atomspace without the source (You hopefully get what I mean) Sorry if I am bad at describing what I mean
I don't know the specifics of that but I assume that it stores it as someone's belief rather than its own. Then PLN can in principle modifies the system's belief, assuming for instance that the person holding the believe is known to be trusted. I don't think it's been tried though.
Nil
Yeah, I've seen that diagram :) I posted one about how I am planning to implement this chatbot however it didn't post correctly, here it is (Hopefully it works this time) Thanks <https://lh3.googleusercontent.com/-qlOvGV3itDY/WXIIFZrb1SI/AAAAAAAAAA4/bfLrDHeX7IIutEf0GXgpucWg2sgmR0m7ACLcBGAs/s1600/PIUP.png> Three short comments: 1) PLN is a set of rules that run on top of the rule engine. Those rules, when triggered, alter the "truth values" on atoms in certain ways. You (or someone) should be able to create a different set of formulas for truth values, that emulate the NARS formulas. You could then run PLN deductions side by side with NARS deductions, and see which you like better (or even create yet other kinds of formulas) 2) we have "relex2logic" rules, which convert natural language to logic expressions. These are ... time-consuming to write, and by nature are incomplete and buggy -- because natural language is a fluid, changing thing. Hand-writing rules is a good exercise, but madness lies in that direction. 3) the current diagram of how things connect is kind-of-ish reviewed here: https://github.com/opencog/opencog/blob/master/opencog/eva/architecture/embodiment.pdf <https://github.com/opencog/opencog/blob/master/opencog/eva/architecture/embodiment.pdf> --linas On Thu, Jul 20, 2017 at 6:14 AM, <[email protected] <javascript:>> wrote: Hi, I am new to Opencog and hopefully building a chatbot with it soon, I was going to use OpenNARS but after some reading I am convinced that PLN > NARS So here are my questions: 1 I am thinking of writing rules that convert input text to a understandable format for Opencog, I am aware that this has already been done with the NLP pipeline However from what I've been told rules that for example convert "that" into what is meant with "that" aren't implemented yet. Example: Tim: My computer keeps crashing Tim: I hate that This rule would turn this into : Tim said that Tim hates that his computer keeps crashing (In atomese) This is just one example of such a rule. 2 I should've asked this before but are such rules even helpful? And is atomese even the language in which such things would be inputted to the system? I am extrapolating from OpenNARS in which input like this is done in Narsese but I am not sure how that works in Opencog yet... 3 What does MOSES do exactly? And is it needed for a chabot? (I know what MOSES does with the evolution of mini programs, I am just not quite sure on what those mini programs do..) That's all of the questions I could think of right now, not sure if I missed one (hopefully not) The chatbot I am building will use an up board which will run Opencog, that should be okay right? http://up-shop.org/up-boards/62-up-board-4b-32-gb-emmc-memory.html <http://up-shop.org/up-boards/62-up-board-4b-32-gb-emmc-memory.html> Thanks! -- You received this message because you are subscribed to the Google Groups "opencog" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected] <javascript:>. To post to this group, send email to [email protected] <javascript:>. Visit this group at https://groups.google.com/group/opencog <https://groups.google.com/group/opencog>. To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/d4fa76b4-80d2-4ea6-96af-e072237b03fd%40googlegroups.com <https://groups.google.com/d/msgid/opencog/d4fa76b4-80d2-4ea6-96af-e072237b03fd%40googlegroups.com?utm_medium=email&utm_source=footer>. For more options, visit https://groups.google.com/d/optout <https://groups.google.com/d/optout>. -- You received this message because you are subscribed to the Google Groups "opencog" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected] <mailto:[email protected]>. To post to this group, send email to [email protected] <mailto:[email protected]>. Visit this group at https://groups.google.com/group/opencog. To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/b8288b31-9201-42e2-b979-a101541639c2%40googlegroups.com <https://groups.google.com/d/msgid/opencog/b8288b31-9201-42e2-b979-a101541639c2%40googlegroups.com?utm_medium=email&utm_source=footer>. For more options, visit https://groups.google.com/d/optout.
-- You received this message because you are subscribed to the Google Groups "opencog" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/opencog. To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/769ad47e-ee0c-9118-88c9-4276cfe53062%40gmail.com. For more options, visit https://groups.google.com/d/optout.
