# US 2016/0253671 A1 · Efficient tail calculation to exploit data correlation

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2016/0253671 A1
- Document ID: US20160253671A1
- Status: Published application
- Kind: A1
- Application number: 14/634,285
- Filed: 2015-02-27
- Published: 2016-09-01
- Earliest claimed: 2015-02-27
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Robert J. Blainey; Barnaby Dalton; Louis Ly; James A. Sedgwick; Lior Velichover; Kai-Ting A. Wang
- Web page: https://patentgrep.com/patent/US20160253671A1

## CPC

- G06Q 20/4016: {involving fraud or risk level assessment in transaction processing}
- G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)}
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A computing system includes at least one processor and at least one module operable by the at least one processor to calculate a tail of a first dataset by determining elements of the first dataset that fall outside of a specified percentile, and determine locations of the first dataset at which elements of the first dataset that fall outside of the specified percentile are located. The at least one module may be operable to calculate a tail of a second dataset by populating a data structure with elements of the second dataset that correspond to the locations of the first dataset, and determining, using the data structure, elements of the second dataset that fall outside of the specified percentile. The at least one module may be operable to output an indication of at least one of the tail of the first dataset or the tail of the second dataset.

## Claims

### Claim at position 1

1-8. (canceled)

### Claim 9

A computing system comprising:
at least one processor; and
at least one module operable by the at least one processor to:
calculate a tail of a first dataset, wherein calculating the tail of the first dataset includes determining elements of the first dataset that fall outside of a specified percentile;
determine locations of the first dataset at which the elements of the first dataset that fall outside of the specified percentile are located;
calculate a tail of a second dataset, wherein calculating the tail of the second dataset includes:
populating a data structure with elements of the second dataset that are located at locations of the second dataset that correspond to the locations of the first dataset, and
determining, using the data structure, elements of the second dataset that fall outside of the specified percentile; and
output an indication of at least one of the tail of the first dataset or the tail of the second dataset. Preliminary Amendment

### Claim 10 (depends on 9)

The computing system of claim 9, wherein determining the elements of the second dataset that fall outside of the specified percentile comprises:
for each particular element in the second dataset other than the elements stored in the data structure:
comparing the particular element with each element in the data structure; and
responsive to determining that the particular element is larger than any element in the data structure, replacing a largest element in the data structure with the particular element.

### Claim 11 (depends on 9)

The computing system of claim 9, wherein the data structure comprises a first data structure, and wherein determining the elements of the first dataset that fall outside of the specified percentile comprises:
determining, based on the specified percentile and on a number of elements that are in the first dataset, a number, m, that indicates how many elements of the first dataset will fall outside of the specified percentile;
storing, in a second data structure, m elements of the first dataset; and
for each particular element in the first dataset other than the m elements stored in the second data structure:
comparing the particular element with each element in the second data structure;
and
responsive to determining that the particular element is larger than any element in the second data structure, replacing a largest element in the second data structure with the particular element.

### Claim 12 (depends on 9)

The computing system of claim 9, wherein the first dataset and the second dataset are both associated with a potential financial transaction, and wherein the at least one module is further operable by the at least one processor to:
determine, based at least in part on the tail of the second dataset, a value of a risk measure associated with the potential financial transaction,
wherein the at least one module operable to output the indication of at least one of the tail of the first dataset or the tail of the second dataset is operable by the at least one processor to output, based at least in part on the value of the risk measure, a recommendation associated with the potential financial transaction.

### Claim 13 (depends on 9)

The computing system of claim 9, wherein the first dataset represents a first point in time for a Monte Carlo simulation, and wherein the second dataset represents a second point in time for the Monte Carlo simulation, the second point in time following the first point in time in the Monte Carlo simulation. Preliminary Amendment

### Claim 14 (depends on 9)

The computing system of claim 9, wherein the first dataset and second dataset are each included in a plurality of sequential datasets, wherein the second dataset occurs subsequent to the first dataset in the plurality of sequential datasets, wherein the data structure comprises a first data structure, and wherein the at least one module is further operable by the at least one processor to:
calculate a tail of a third dataset in the plurality of sequential datasets without using locations of any datasets prior to the third dataset in the plurality of sequential datasets, wherein the third dataset occurs subsequent to the second dataset in the plurality of sequential datasets, and wherein the at least one module operable to calculate the tail of the third dataset is operable to determine elements of the third dataset that fall outside of the specified percentile;
determine locations of the third dataset at which the elements of the third dataset that fall outside of the specified percentile are located;
calculate a tail of a fourth dataset, wherein the fourth dataset occurs subsequent to the third dataset in the plurality of sequential datasets, and wherein the at least one module operable to calculate the tail of the fourth dataset is operable to:
populate a second data structure with elements of the fourth dataset that are located at locations of the fourth dataset that correspond to the locations of the third dataset, and
determine, using the second data structure, elements of the fourth dataset that fall outside of the specified percentile; and
output an indication of at least one of the tail of the third dataset or the tail of the fourth dataset.

### Claim 15 (depends on 9)

The computing system of claim 9, wherein the data structure comprises a heap-like data structure.

### Claim 16 (depends on 9)

The computing system of claim 9, wherein the data structure comprises a min-heap or a max-heap.

### Claim 17

A computer program product comprising a computer readable storage medium having stored thereon:
program instructions executable by at least one processor to cause at least one computing device to calculate a tail of a first dataset, wherein calculating the tail of the first dataset includes determining elements of the first dataset that fall outside of a specified percentile;
program instructions executable by the at least one processor to cause the at least one computing device to determine locations of the first dataset at which the elements of the first dataset that fall outside of the specified percentile are located;
program instructions executable by the at least one processor to cause the at least one computing device to calculate a tail of a second dataset, wherein calculating the tail of the second dataset includes:
populating a data structure with elements of the second dataset that are located at locations of the second dataset that correspond to the locations of the first dataset, and
determining, using the data structure, elements of the second dataset that fall outside of the specified percentile; and
program instructions executable by the at least one processor to cause the at least one computing device to output an indication of at least one of the tail of the first dataset or the tail of the second dataset.

### Claim 18 (depends on 17)

The computer program product of claim 17, wherein determining the elements of the second dataset that fall outside of the specified percentile comprises:
for each particular element in the second dataset other than the elements stored in the data structure:
comparing the particular element with each element in the data structure; and
responsive to determining that the particular element is smaller than any element in the data structure, replacing a smallest element in the data structure with the particular element.

### Claim 19 (depends on 17)

The computer program product of claim 17, wherein the data structure comprises a first data structure, and wherein determining the elements of the first dataset that fall outside of the specified percentile comprises:
determining, based on the specified percentile and on a number of elements that are in the first dataset, a number, m, that indicates how many elements of the first dataset will fall outside of the specified percentile;
storing, in a second data structure, m elements of the first dataset; and
for each particular element in the first dataset other than the m elements stored in the second data structure:
comparing the particular element with each element in the second data structure;
and
responsive to determining that the particular element is smaller than any element in the second data structure, replacing a smallest element in the second data structure with the particular element.

### Claim 20 (depends on 17)

The computer program product of claim 17, wherein the first dataset and second dataset are each included in a plurality of sequential datasets, wherein the second dataset occurs subsequent to the first dataset in the plurality of sequential datasets, wherein the datastructure comprises a first data structure, and wherein the computer readable storage medium has further stored thereon:
program instructions executable by the at least one processor to cause the at least one computing device to calculate a tail of a third dataset in the plurality of sequential datasets without using locations of any datasets prior to the third dataset in the plurality of sequential datasets, wherein the third dataset occurs subsequent to the second dataset in the plurality of sequential datasets, and wherein calculating the tail of the third dataset includes determining elements of the third dataset that fall outside of the specified percentile;
program instructions executable by the at least one processor to cause the at least one computing device to determine locations of the third dataset at which the elements of the third dataset that fall outside of the specified percentile are located;
program instructions executable by the at least one processor to cause the at least one computing device to calculate a tail of a fourth dataset, wherein the fourth dataset occurs subsequent to the third dataset in the plurality of sequential datasets, and wherein calculating the tail of the fourth dataset includes:
populating a second data structure with elements of the fourth dataset that are located at locations of the fourth dataset that correspond to the locations of the third dataset, and
determining, using the second data structure, elements of the fourth dataset that fall outside of the specified percentile; and
program instructions executable by the at least one processor to cause the at least one computing device to output an indication of at least one of the tail of the third dataset or the tail of the fourth dataset.
