Extracting regression rules from neural networks
Neural Networks
Discovering Polynomials to Fit Multivariate Data Having Numeric and Nominal Variables
Progress in Discovery Science, Final Report of the Japanese Discovery Science Project
Hi-index | 0.00 |
The World Wide Web is a complex system that changes over time. However, like any dynamic system, we can attempt to understand it in scientific terms and describe regularities in its behavior. To this end, we have collected measurements for a number of Web-related variables, including the daily changes in Web pages. To the resulting data, we have applied RF6, an algorithm that discovers descriptive laws in the form of polynomial expressions that hold under conditions on nominal variables. Applying this technique to the collected data has produced some initial laws of Web dynamics that encourage further research along these lines.