Data processing: artificial intelligence – Machine learning
Patent
1997-05-20
1999-04-27
Downs, Robert W.
Data processing: artificial intelligence
Machine learning
706 16, G06F 1518
Patent
active
058976272
ABSTRACT:
A method including providing a database with a plurality of items, each item having several attributes, and developing and displaying a cluster map (SOM) attribute plane from the items and attributes, including clusters of items having similar attributes. Selecting an attribute of interest and splitting the plurality of items into clusters of interest based on distribution of the selected attribute. Utilizing rule induction to provide a rule for each cluster of interest, drawing a line around adjacent items in the SOM attribute plane for each rule to form at least one complete fenced area in the display, and comparing fenced areas to clusters to separate statistically meaningful rules from statistically meaningless rules or coincidences.
REFERENCES:
patent: 5210868 (1993-05-01), Shimada et al.
patent: 5396580 (1995-03-01), Fu
patent: 5687286 (1997-11-01), Bav-Yam
patent: 5720007 (1998-02-01), Hekmatpour
patent: 5724573 (1998-03-01), Agrawal et al.
Gardner Robert M.
Leivian Robert H.
Downs Robert W.
Motorola Inc.
Parsons Eugene A.
LandOfFree
Method of determining statistically meaningful rules does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of determining statistically meaningful rules, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of determining statistically meaningful rules will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-682949