Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1998-05-07
2001-07-10
Amsbury, Wayne (Department: 2171)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000, C707S793000
Reexamination Certificate
active
06260036
ABSTRACT:
TECHNICAL FIELD
This invention relates to a method and apparatus for organizing and retrieving data in a parallel transaction data base.
DESCRIPTION OF THE PRIOR ART
Recently, the importance of database mining is growing a rapid pace by the increasing use of computing for various applications. Progress in bar-code technology has made it possible for retail organizations to collect and store massive amounts of sales data. Catalog companies can also collect sales data from the orders they received. A record in such data typically consists of the transaction date, the items bought in that transaction, and possibly the customer-id if such a transaction is made via the use of a credit card or customer card.
The self-organizing map (SOM) [T. Kohonen, The
Self-Organizing Map, Proc. IEEE,
vol. 73, pp. 1551-1558, 1985; T. Kohonen,
Self-Organizing Maps,
Springer, 1995] is a neural network model that is capable of projecting high-dimensional input data onto a low-dimensional (typically two-dimensional) array. This nonlinear projection produces a two-dimensional “feature-map” that can be useful in detecting and analyzing features in the input space. SOM techniques have been successfully applied in a number of disciplines including speech recognition [T. Kohonen, The neural phonetic typewriter,
Computer,
21 (3), pp. 11-22, 1988], image classification [S. Lu, Pattern classification using self-organizing feature maps, in
IJCNN International Joint Conference on Neural Networks,
Newport Beach, Calif. February 1994], and document clustering [K. Lagus, T. Honkela, S. Kaski, and T. Kohonen, Self-organizing maps of document collections: a new approach to interactive exploration, in
Proc. Second Intl. Conf. On Knowledge Discovery and Data Mining,
Portland, 238-243, August, 1996]. An extensive bibliography of SOM applications is given in T. Kohonen, The
Self-Organizing Map, Proc. IEEE,
vol. 73, pp. 1551-1558, 1985 and is also available at T. Kohonen, J. Hynninen, J. Kangas, and J. Laaksonen, SOMPAK: The self-organizing map program package, Helsinki University of Technology, http:/
ucleus.hut.fi
nrc/som_pak.
Neural networks are most often used to develop models that are capable of predicting or classifying an output as a response to a set of inputs to the trained network. Supervised learning is used to train the network against input data with known outputs. In contrast, the SOM typically is applied to data in which specific classes or outcomes are not known apriori, and hence training is done in unsupervised mode. In this case, the SOM can be used to understand the structure of the input data, and in particular, to identify “clusters” of input records that have similar characteristics in the high-dimensional input space. Now input records (with the same dimensionality as the training vectors) can be analyzed (and assigned to clusters) using the neural weights computed during training. An important characteristic of the SOM is the capability to produce a structured ordering of the input vectors. This “self-organization” is particularly useful in clustering analysis since it provides additional insight into relationship between the identified clusters.
REFERENCES:
patent: 5675819 (1997-10-01), Schuetze
patent: 5706495 (1998-01-01), Chadha et al.
patent: 5778362 (1998-07-01), Deerwester
patent: 5802525 (1998-09-01), Rigoutsos
patent: 5832182 (1998-11-01), Zhang et al.
patent: 5857179 (1999-01-01), Vaithyanathan et al.
patent: 5864855 (1999-01-01), Ruocco et al.
patent: 5940825 (1999-08-01), Castelli et al.
patent: 5983224 (1999-11-01), Singh et al.
patent: 6003029 (1999-12-01), Agrawal et al.
patent: 6012058 (2000-01-01), Fayyad et al.
patent: 6078918 (2000-06-01), Allen et al.
Ng et al., “Vector Quantization for Lossless Textual Data Compression”, IEEE, 1995, pp. 474.*
Percival, “Compressed Representation of Backscatter Ionogram database using Kahunen-Loeve Techniques”, IEEE, 1995, pp. 574-578.*
Ng et al., “Relational Database Compression Using Augmented Vector Quantization”, IEEE, 1995, pp. 540-549.
Almasi George S.
Lawrence Richard Douglas
Rushmeier Holly Edith
Amsbury Wayne
Pardo Thuy
LandOfFree
Scalable parallel algorithm for self-organizing maps with... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Scalable parallel algorithm for self-organizing maps with..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Scalable parallel algorithm for self-organizing maps with... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2514471