Hi BW, On Jan 14, 2012, at 9:59 PM, B W wrote:
> The first critical step in determining whether or not OODT should be > adopted seems to be predicated on being able to define a Common Data > Dictionary (CDD) for a given domain or at least be able to map an internal > Synonymous Data Dictionary to the Common Data Dictionary. Yep this is super important both on the File Manager side (and data processing), as well as on the information integration side when integrating a local site's data model into the overall common data model. > > I think I read in a published OODT experience report that the task > required an individual with at least 2 years experience. > > Is there a consensus on what is required across the board? To each their own. I'd say it's definitely more difficult to do without any prior OODT or data systems experience, but not impossible. > > Also, have there been any AI approaches applied such as clustering/combined > weighted algorithms to aid in the task? We've tried that approach and found it to be kind of garbage in, garbage out, since most of the work is in semantic mapping and so forth which typically requires someone (a "human") that knows what they are doing there. My 2c. Cheers, Chris ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Chris Mattmann, Ph.D. Senior Computer Scientist NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA Office: 171-266B, Mailstop: 171-246 Email: [email protected] WWW: http://sunset.usc.edu/~mattmann/ ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Adjunct Assistant Professor, Computer Science Department University of Southern California, Los Angeles, CA 90089 USA ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
