On Tue, 28 Nov 2000 11:09:39 Brian Bray wrote: ... >A specific study at a specific site >will need to customize their mining tools to the available data. I.E. >the mining tools will themselves reference the same meta-information. In >the end, these tools would format the information for conventional >databases and/or statistical analysis packages. Hi Brian, The OIO is extending the "plug-and-play" metadata concept to the data mining tools. We will be providing "plug-and-play" equations and data manipulation routines (e.g. Linkers-metadata). In the latest release OIO-0.9.4, we have the beginnings of this capability in "merging" data across Forms (by item name). So, while you are correct that we can output the data for external analysis, the holy grail is to have customizable and *reusable* data mining modules (OIO Reports) as well as dynamic data structures (OIO Forms). Just like the Forms that can be indexed, searched, and retrieved from online libraries, the idea is to re-use these data mining tools. As far as "customization" of the mining tools is concerned, this will occur at 2 different levels: 1) changing the tools themselves - i.e. changing an equation, 2) applying them to different sets of data - analyzing cancer survival rather than cardiac event survival. Current statistical packages allow both of these (easier to do #2 than #1). The difference OIO brings is ease of doing #1 (how many people can write a SAS routine?) and integration within a full-featured data collection/management system - so export/import of data is not necessary. (Again, how many people can correctly export/import data into a statistical package?) >At least this was the theory for Circare! It really remains to be seen >what the advantages and disadvantages are in practice and what tools are >most useful for this use case. We have been using the OIO with a team of four clinicians and a secretary at Harbor-UCLA. We started with the Forms design and then data collection around Feb 2000. We started analyzing data in July 2000. The biggest advantage is clinician involvement through the fostering of a "collaborative" attitude. The system serves their needs and is not seen as something that is *imposed* to serve some "higher" purpose. We have over 1200 patients in the system and all the users agree that it is easier to use the system than to 1) do it by hand, or 2) use Microsoft Excel/Access. In the beginning, they liked the web-interface but were suspicious that we can generate useful reports from the data. Since we implemented our first data mining tool last month, the enthusiasm for using the system sky-rocketed. ... >Also, its quite clear that the >schemas/archetypes/DTDs used in a particular context are not going to be >developed completely independently. Similarities between type >definitions can be used to reduce the complexity of the mining task. Exactly. OIO takes advantage of this to allow merging of data across Forms/versions. The creation of "Linkers"-metadata will take this one step further and allow arbitrary translations between "type definitions"=OIO Forms/Items. Once in place, these "Linkers" will allow inter-operation between different data sets that exist in different term-spaces. ... >Going back to your original comments, Another way of looking at this is >that these systems *do* have, in practice, a "centrally defined data >structure and term glossary.", it's just not hard coded into the >application code and it's defined locally. I still wouldn't call it "central". How about "Locally-defined data structure and terms that can be pooled and shared globally"? A bottom-up data structure/terms, in other words, ="democratic". Andrew --- Andrew P. Ho, M.D. OIO: Open Infrastructure for Outcomes www.TxOutcome.Org Assistant Clinical Professor Department of Psychiatry, Harbor-UCLA Medical Center University of California, Los Angeles Join 18 million Eudora users by signing up for a free Eudora Web-Mail account at http://www.eudoramail.com
