| US 7,580,848 B2 | ||
| Method of and system for analyzing, modeling and valuing elements of a business enterprise | ||
| Jeff Scott Eder, Mill Creek, Wash. (US) | ||
| Assigned to Asset Trust, Inc., Bothell, Wash. (US) | ||
| Filed on Oct. 30, 2002, as Appl. No. 10/283,083. | ||
| Application 10/283083 is a continuation of application No. 09/938874, filed on Aug. 27, 2001. | ||
| Application 09/938874 is a continuation of application No. 08/999245, filed on Dec. 10, 1997, abandoned. | ||
| Application 08/999245 is a continuation of application No. 08/779109, filed on Jan. 06, 1997, granted, now 6,393,406. | ||
| Prior Publication US 2008/0215386 A1, Sep. 04, 2008 | ||
| Int. Cl. G06Q 40/00 (2006.01) | ||
| U.S. Cl. 705—7 [705/35] | 84 Claims |
| 1. A computer-implemented network model method, comprising:
receiving first input data into an initial network model to develop an initial model configuration;
receiving an initial input data set from said initial model configuration and a second input data as inputs into a second
model stage to develop and output an improvement to the initial input data set by evolving a plurality of network models using
a plurality of genetic algorithms that exchange a plurality of data between successive generations of two or more independent
subpopulations, said second input data comprising one of said first input data, data not included in said first input data,
and a combination thereof;
receiving said second model stage output as an input into third, induction stage in order to refine and summarize the data
selection in the second stage model output and output a refined input data set, and
receiving said induction stage output as an input into a fourth model stage to develop and output a final network model by
training a network model using the refined input data set where all input data represents a physical object or substance,
and
where said final network model supports a regression analysis.
|