Robert,
As I said in my previous short message to the list, Jess is entirely
appropriate for your application. I immediately grasped, from the
content of your original message, that you would be processing a
database of complete records of (possibly preexisting) data, and
wanted to validate or possibly correct it, remove duplicate records,
etc. This process has recently been referred to as "data cleansing" and
is often done as a preliminary step to data mining. ESs work very well
in many cleansing and mining applications, both.
I think Robert Quillen wrote:
[Charset iso-8859-1 unsupported, filtering to ASCII...]
> Perhaps I should explain my application a bit more. The example constraints
> I gave were rather simple.
>
> We write software for life insurance companies. Our customers want to sell
> insurance policies over the web. This boils down to collecting and
> verifying an insurance applicant's data.
>
> The data will generally be collected via a set of forms. The solution must
> be form technology neutral. We know that there will be both PDF and HTML,
> but there could be something else that we haven't thought of yet.
>
> Our customers like "fine-grained" forms for data collection. They generally
> only have 2 or 3 fields with lots of descriptive text. This means there are
> generally a hundred or so forms for a typical application.
>
> The form workflow is dynamic. The user only sees the ones that are
> appropriate, based on previously collected data. (i.e. The tobacco usage
> forms are only presented if the applicant said they use tobacco on an
> earlier form.)
>
> The data can also be provided from external sources. Some sources include
> contact management/sales support systems, customer databases, legacy apps,
> etc.
>
> The data validation and completeness checking has to be done separate from
> the form because:
> 1. We don't write the forms.
> 2. We don't want to write the forms. :)
> 3. There can be dependencies between the data that cross multiple forms.
> The rules for checking the data on form B can be dependent upon the data
> collect by form A.
> 4. The data can come from external sources beyond our control.
>
> Even if we did code the forms ourselves, numbers 3 and 4 would be big
> problems.
>
> So. I need to write a component that can analyze a set of data and answer
> two questions:
> 1. "Is this a complete life insurance application?"
> 2. "Is this subset of the application complete?"
>
> Number 2 is needed for the dynamic form workflow requirement. Since I know
> what fields are on each form, I can use that information to decide if a
> particular form should be displayed.
>
>
> So now do you think Jess is appropriate?
>
> -Robert Quillen
---------------------------------------------------------
Ernest Friedman-Hill
Distributed Systems Research Phone: (925) 294-2154
Sandia National Labs FAX: (925) 294-2234
Org. 8920, MS 9012 [EMAIL PROTECTED]
PO Box 969 http://herzberg.ca.sandia.gov
Livermore, CA 94550
---------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the
list (use your own address!) List problems? Notify [EMAIL PROTECTED]
---------------------------------------------------------------------
