Fuzzy Rank

There are a number of ways to select the best set of inputs to be used to form a regression for a particular output. Simply crossplotting each input against the output can give an indication of the quality of linear or multiple linear regression models that could be formed. In the development of an on line Expert system for oil prospecting we have encountered many non-linear or complex problems. To address this software was developed based on a single stage fuzzy-ranking algorithm to select inputs best suited for predicting the desired output. This global algorithm statistically determines how well a particular input could resolve a particular output with respect to any number of other inputs using fuzzy curve analysis.

In order to eliminate the interpretive bias of manually examining fuzzy curves a second stage algorithm was written. That software is now available on line, with an optional Windows useable executable for installation on your computer.

Introduction Instructions on organizing and formatting your data for use with FuzzyRank, as well as interpreting results.

Basic Theory A short essay on the theory behind the algorithm.

Log on to Fuzzy Rank A password protected account is required to use Fuzzy Rank. E-mail Principal Investigator Robert Balch, or call (575) 835-5305 for information.