Software has been granted a patent for a data imputation computer system that automatically corrects and completes data in sparse datasets. The system divides records into groups, identifies properties to be ignored, and predicts values to fill in empty data. This innovative technology enhances data accuracy and completeness. GlobalData’s report on Software gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Software, was a key innovation area identified from patents. Software's grant share as of February 2024 was 77%. Grant share is based on the ratio of number of grants to total number of patents.
Data imputation system for correcting and completing sparse datasets
A recently granted patent (Publication Number: US11875234B2) discloses a data imputation computer system designed to correct and complete datasets by predicting missing values. The system, comprising an interface, processor, and memory, operates by receiving a sparse dataset organized into classes with associated properties. It then divides the data records into groups based on similar property values, identifies properties to be ignored within each group, further divides the groups into sub-groups, predicts values for empty entries, applies these predictions, and outputs the corrected dataset. The system utilizes clustering algorithms to generate groups and sub-groups, with the first and second clustering algorithms being distinct from each other. Additionally, the system calculates property scores, weights distances among property values, and determines which properties to ignore based on predefined thresholds.
Furthermore, the patent includes a data imputation method and a non-transitory computer-readable storage medium storing instructions for dataset correction and completion. The method and storage medium follow a similar process as the system, involving the division of data records into groups and sub-groups, identification of properties to be ignored, prediction of missing values, and output of the corrected dataset. The method and storage medium also incorporate clustering algorithms, property score calculations, and the selection of properties to be ignored based on predetermined thresholds. Overall, the patented technology offers a comprehensive solution for imputing missing data in datasets, enhancing data accuracy and completeness for various applications.
To know more about GlobalData’s detailed insights on Software, buy the report here.
Premium Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.