公开(公告)号：US11704388B2

公开(公告)日：2023-07-18

申请号：US17862510

申请日：2022-07-12

Applicant: SAS Institute Inc.

Inventor： Christian Macaro ,  Fedor Reva ,  Rocco Claudio Cannizzaro

IPC: G06F17/11 ,  G06F18/2413 ,  G06Q10/0635 ,  G06F18/2411 ,  G06F18/23213

CPC classification number: G06F18/24137 ,  G06F17/11 ,  G06F18/23213 ,  G06F18/2411 ,  G06Q10/0635

Abstract: A computing device determines a disaggregated solution vector of a plurality of variables. A first value is computed for a known variable using a predefined density distribution function, and a second value is computed for an unknown variable using the computed first value, a predefined correlation value, and a predefined aggregate value. The predefined correlation value indicates a correlation between the known variable and the unknown variable. A predefined number of solution vectors is computed by repeating the first value and the second value computations. A solution vector is the computed first value and the computed second value. A centroid vector is computed from solution vectors computed by repeating the computations. A predefined number of closest solution vectors to the computed centroid vector are determined from the solution vectors. The determined closest solution vectors are output.

公开(公告)号：US20230185883A1

公开(公告)日：2023-06-15

申请号：US17862510

申请日：2022-07-12

Applicant: SAS Institute Inc.

Inventor： Christian Macaro ,  Fedor Reva ,  Rocco Claudio Cannizzaro

IPC: G06K9/62 ,  G06Q10/06

CPC classification number: G06K9/6272 ,  G06K9/6223 ,  G06K9/6269 ,  G06Q10/0635

Abstract: A computing device determines a disaggregated solution vector of a plurality of variables. A first value is computed for a known variable using a predefined density distribution function, and a second value is computed for an unknown variable using the computed first value, a predefined correlation value, and a predefined aggregate value. The predefined correlation value indicates a correlation between the known variable and the unknown variable. A predefined number of solution vectors is computed by repeating the first value and the second value computations. A solution vector is the computed first value and the computed second value. A centroid vector is computed from solution vectors computed by repeating the computations. A predefined number of closest solution vectors to the computed centroid vector are determined from the solution vectors. The determined closest solution vectors are output.

公开(公告)号：US11016871B1

公开(公告)日：2021-05-25

申请号：US17129536

申请日：2020-12-21

Applicant: SAS Institute Inc.

Inventor： Rocco Claudio Cannizzaro ,  Christian Macaro

IPC: G06F11/30 ,  G06F11/34 ,  G06F9/4401 ,  G06K9/62 ,  G06F9/38 ,  G06F9/48

Abstract: Resource consumption associated with executing a bootstrapping process on a computing device can be reduced. For example, a system can receive a dataset including observations. The system can then instantiate one or more thread objects configured to execute a bootstrapping process that involves multiple iterations. Each iteration can involve: determining a respective set of probabilities based on an observation distribution associated with the dataset, executing a function based on the respective set of probabilities to determine a respective metric value, and storing the respective metric value in memory. This iterative process may be faster and less computationally intensive than traditional bootstrapping approaches. After completing the iterative process, the system may access the memory to obtain the metric values, determine a distribution of metric values based on at least some of the metric values, and store the distribution of metric values in the memory for further use.