KnowledgeMiner, a “data mining” tool, has been updated to version 4.0, adding at least one Mac only feature.
The application is a neural network analysis program that takes inputs and results for systems and derives equations that describe the results in terms of the inputs. Although not designed specifically for chemistry it allows analysis of effects of conditions on chemical reactions.
Version 4.0 extends AppleScript support and facilitates workflow processing through the importing data from databases or spreadsheets, data preprocessing, data mining, prediction/classification of new data and returning processed data. New features include creating and predicting from within a script, making documents of a certain table dimension via AppleScript and calculation of the ROC integral as a measure of classification.
KnowledgeMiner seems especially Mac friendly (note its AppleScript enhancements). And on the frequently asked questions page, it notes: “With Motorola’s new math libraries and the higher clock speeds of their chips, math intensive applications such as KnowledgeMiner are best run on a Mac; Byte’s recent benchmarks show the new Macs running twice as fast in SpecInt and 50 percent faster in SpecFPU than comparable P5 or P6 chips running at the same speed.”
Included datasets and examples include; stock market trends, medical diagnosis, global temperature predictions, failure of materials (like the Challenger Space Shuttle O-Ring), wine recognition and national economy. Prices range from US$65 for the student version up to $780 for the gold edition. For more information on KnowledgeMiner go to the product Web site.