Distribution theory based enrichment of sparse data for machine learning
AssigneeHoneywell International Inc.
InventorRavindra K. Shetty
Life of this patent
Estimating the term…
Abstract
A technique for enriching sparse data for machine learning techniques such as supervised artificial neural network includes receiving the sparse data and enriching the received data around a deviation of the mean of the received data using a predetermined distribution. The technique further includes outputting the enriched data for unbiased and increased performance during the machine learning.
Claims52
A computer-implemented method for enriching sparse data for machine learning, comprising:
- receiving the sparse data, wherein the received data comprises data selected from the group consisting of static data and real-time data;
- if the received data is static data, then reading a sample of the received static data using a predetermined window length;
- and
- if the received data is real-time data, then reading a sample of the received real-time data using a dynamically varying window of predetermined window length;
- enriching the received data around a deviation of the mean of the received data using a predetermined distribution;
- and
- outputting the enriched data for unbiased learning and improved performance during the machine learning.
A computer-implemented method for enriching sparse data for machine learning, comprising:
- receiving the sparse data;
- enriching the received data around a deviation of the mean of the received data using a predetermined distribution, wherein the predetermined distribution comprises distributions selected from the group consisting of normal distribution, exponential distribution, logarithmic distribution, chi-square distribution, t-distribution, and F-distribution;
- and
- outputting the enriched data for unbiased learning and improved performance during the machine learning.
A computer readable medium having computer-executable instructions for performing a method of machine learning when only sparse data is available, comprising:
- enriching the sparse data around a deviation of the mean of the received data using a predetermined distribution selected from the group consisting of normal distribution, exponential distribution, logarithmic distribution, chi-square distribution, t-distribution, and F-distribution;
- and
- outputting the enriched data for unbiased machine learning.
A computer system for a machine learning in a sparse data environment, comprising:
- a storage device;
- an output device;
- and
- a processor programmed to repeatedly perform a method, comprising:
- receiving the data;
- enriching the received data around a deviation of mean of the received data using a predetermined distribution selected from the group consisting of normal distribution, exponential distribution, logarithmic distribution, chi-square distribution, t-distribution, and F-distribution;
- and
- outputting the enriched data for unbiased machine learning.
A computer-implemented system for machine learning in a sparse data environment, comprising:
- a receive module to receive sparse data;
- a database coupled to the receive module to receive and store sparse data;
- and
- a unique numeric transformation module coupled to the database to extract words from text stored in the database and to transform each of the extracted words into a unique numerical representation;
- an analyzer to enrich the received data around a deviation of the received data using a predetermined distribution;
- and
- an output module coupled to the analyzer to output the enriched data for unbiased learning and increased performance during machine learning.
A computer-implemented system for machine learning in a sparse data environment, comprising:
- a receive module to receive sparse data, wherein the received data comprises data selected from the group consisting of static data and real-time data;
- a reading module coupled to the receive module that reads a sample of the received data having a predetermined window length;
- an analyzer to enrich the received data around a deviation of the received data using a predetermined distribution;
- and
- an output module coupled to the analyzer to output the enriched data for unbiased learning and increased performance during machine learning.
Dependencies are inferred from the claim text.