- From: William Waites <ww@styx.org>
- Date: Sun, 26 Sep 2010 21:57:11 +0100
- To: Bob Ferris <zazi@elbklang.net>
- CC: semantic-web@w3.org
- Message-ID: <4C9FB3A7.8060007@styx.org>
On 10-09-23 19:43, Bob Ferris wrote: > However, these simple statements do not say anything about the levels or > weightings of these cognitive patterns of this person, rather then > something about the related activity, e.g. playing or watching, or > characteristic dynamics. Almost want to write down something like, ## alice thinks she knows a lot about football { alice believes { alice cco:expertise <Football> } } weight 0.9 . ## alice thinks bob is a pretty good football player { alice believes { bob cco:skill <Football> } } weight 0.8 . ## mary thinks alice is not quite so expert as she claims { mary believes { alice cco:expertise <Football> } } weight 0.6 . A nice feature of this approach is I can refer to it in a nested way which keeps the scope clear, { ww believes G1 } weight 0.2 . where G1 is the example above - i.e. I think the example is mostly nonsense. On the basis of this I might come up with, { ww believes { bob cco:skill <Football> } } weight 0.12 . or, I am not particularly confident at all that bob is a good football player on the basis of the facts available [*] For the specific of Bob's using CognitiveCharacteristics, is it true that the weight goes on the characteristic itself or does it make more sense in terms of belief about the characteristic? Just some ideas anyways, not immediately practical... Cheers, -w [*] average expertise of alice, alice_exp = (0.9 + 0.6)/2 = 0.75 skill of bob, skill_bob = 0.8 * alice_exp = 0.6 normalised to our confidence in things that can be inferred from the knowledge base, skill_bob_norm = skill_bob * 0.2 = 0.12 -- William Waites <ww@styx.org> Mob: +44 789 798 9965 Fax: +44 131 464 4948 CD70 0498 8AE4 36EA 1CD7 281C 427A 3F36 2130 E9F5
Received on Sunday, 26 September 2010 20:59:37 UTC