Enhancing knowledge discovery from multiple data sets using multiple support vector machines

Abstract

A system and method for enhancing knowledge discovery from data using multiple learning machines in general and multiple support vector machines in particular. Training data for a learning machine is pre-processed in order to add meaning thereto. Pre-processing data may involve transforming the data points and / or expanding the data points. By adding meaning to the data, the learning machine is provided with a greater amount of information for processing. With regard to support vector machines in particular, the greater the amount of information that is processed, the better generalizations about the data that may be derived. Multiple support vector machines, each comprising distinct kernels, are trained with the pre-processed training data and are tested with test data that is pre-processed in the same manner. The test outputs from multiple support vector machines are compared in order to determine which of the test outputs if any represents a optimal solution. Selection of one or more kernels may be adjusted and one or more support vector machines may be retrained and retested. Optimal solutions based on distinct input data sets may be combined to form a new input data set to be input into one or more additional support vector machine.

Description

technical field [0001] The present invention relates to the use of learning machines to discover knowledge from data. More particularly, the present invention relates to the optimization of learning machines and associated input and output data in order to improve knowledge discovered from multiple data sets. Background of the invention [0002] Knowledge discovery is the ideal end product of data collection. Recent advances in database technology have led to an explosion of systems and methods for generating, collecting, and storing enormous amounts of data. While database technology has enabled the efficient acquisition and storage of large data sets, the task of facilitating human understanding of the information in such data has grown even more difficult. With numerous existing technologies, this problem has become intractable. Therefore, there exists a need for a new generation of automatic knowledge discovery tools. [0003] As a concrete example, the Human Genome ...

AI Technical Summary

Problems solved by technology

For example, a significant disadvantage of backpropagation neural networks is that the empirical hazard function can have many local minima, which can easily be obscured by this technique from finding the best solution

Specifically, the skill of the practitioner implementing the neural network determines the ultimate benefit, but even factors as benign as an apparently random choice of initial weights can lead to poor results

Furthermore, gradient-based methods used in neural network learning are historically slow

Another disadvantage: the inverse function has a scaling factor which affects the quality of the approximation

[0010] Additionally, the ability of SVMs to discover knowledge from data is limited by the choice of kernel

Images