Browse Prior Art Database

Method and apparatus for incorporating calculus as a data mining technology Disclosure Number: IPCOM000200740D
Publication Date: 2010-Oct-27
Document File: 1 page(s) / 40K

Publishing Venue

The Prior Art Database


Data mining technologies currently work with absolute values. Our idea extends current methods by allowing data to be selected based on the gradient of the data graph at that point, or based on local maxima and minima.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 100% of the total text.

Page 01 of 1

Method and apparatus for incorporating calculus as a data mining technology

Currently data mining technologies such as SQL work by comparing absolute values which have to be precalculated. This means that A and B have to be known or calculated values. This doesn't allow fine grained control of the results set. Our invention allows business to more easily spot patterns and to gain business intelligence by analysing trends.

    Using SQL as an example. A typical query would be SELECT * FROM sales

history WHERE sales < :sales

   _Here we reply on literal values which have to be precalculated. It is impossible to select if sales are increasing or decreasing and we have no idea what the trends are at the time. Our invention would allow us to select data based on the trends. Our invention would also allow information to be returned on the trends themselves, so the business can tell how strong the trends are at the given data point.

    A typical embodiment of this invention could be illustrated using the example of SQL, but extending the language syntax to allow examples such as
WHERE gradient(field) positive or
WHERE gradient(field) negative or
WHERE local


minimum(field) or

WHERE local


    The field would have to be plottable, we suggest an embodiment which simply replots the graph rather than an embodiment which gets bogged down in calculus on the data.

Another aspect would be
SELECT gradient(field) FROM table