My Photo

« Twitter the tide raises all boats | Main | ACL 2009 Accepted Papers »

April 13, 2009


Robert Butler

I disagree that Fernando has no requirements for the stuff that happens "inside." Indeed, I see that as the very essence of what he is trying to understand. While your "agents" as you mention read and write semantic meaning that are formally specified according to some ontology, Fernando's approach reads and responds to semantic meaning specified in the form of massive amounts of web text. Web text is inherently less formal and imprecise... enter large scale consumption to capture all of the edge cases and better "understand" the inexplicit context. Agents consuming formal ontological statements have the advantage of explicitly defined contexts since they are formally specified. It is not important what is consumed, but the action that can be taken upon consumption (see my blog post for perhaps a better description:

Further, I see the large scale data processing described by Fernando as supporting human cognitive models well. Much of what we are is determined by the large volumes of data we have processed since childhood. There is a very real aspect to social meaning derived from repeated exposure to some "truth." This, in large part, is what is so interesting about the social web. It opens the horizons to what we can experience and understand socially. While I am no expert, I see a fundamental complimentary role between both of these approaches. There are aspects of human cognition that seem extremely difficult, if not impossible to model with a pure descriptive formal specification. At the same time, higher levels of rational thought would be difficult to successfully model by a purely statistical/social approach.

The comments to this entry are closed.

Twitter Updates

    follow me on Twitter

    March 2016

    Sun Mon Tue Wed Thu Fri Sat
        1 2 3 4 5
    6 7 8 9 10 11 12
    13 14 15 16 17 18 19
    20 21 22 23 24 25 26
    27 28 29 30 31    


    Blog powered by Typepad