# US 2021/0342356 A1 · Malicious activity detection system capable of efficiently processing data accessed from databases and generating alerts for display in interactive user interfaces

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

- Number: US 2021/0342356 A1
- Document ID: US20210342356A1
- Status: Published application
- Kind: A1
- Application number: 17/323,284
- Filed: 2021-05-18
- Published: 2021-11-04
- Earliest claimed: 2015-08-28
- Assignees: Palantir Technologies Inc.
- Applicants: Palantir Technologies Inc.
- Inventors: Craig Saperstein; Eric Schwartz; Hongjai Cho
- Web page: https://patentgrep.com/patent/US20210342356A1

## CPC

- G06F 16/24575: {using context}
- G06F 16/2365: {Ensuring data consistency and integrity}
- G06F 16/24544: {Join order optimisation}
- G06F 16/2456: {Join operations}
- G06F 16/248: Presentation of query results
- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06Q 20/4016: {involving fraud or risk level assessment in transaction processing}
- G06Q 40/12: Accounting
- H04L 63/1416: {Event detection, e.g. attack signature detection}
- H04L 63/1425: {Traffic logging, e.g. anomaly detection}
- H04L 63/20: {for managing network security; network security policies in general (filtering policies H04L63/0227)}
- H04L 2463/102: applying security measure for e-commerce
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- 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
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Various systems and methods are provided that retrieve raw data from issuers, reorganize the raw data, analyze the reorganized data to determine whether the risky or malicious activity is occurring, and generate alerts to notify users of possible malicious activity. For example, the raw data is included in a plurality of tables. The system joins one or more tables to reorganize the data using several filtering techniques to reduce the processor load required to perform the join operation. Once the data is reorganized, the system executes one or more rules to analyze the reorganized data. Each rule is associated with a malicious activity. If any of the rules indicate that malicious activity is occurring, the system generates an alert for display to a user in an interactive user interface.

## Claims

### Claim 1

(canceled)

### Claim 2

A computing system comprising:
a database storing first data;
a computer processor; and
a computer readable storage medium storing program instructions configured for execution by the computer processor in order to cause the computing system to:
select a behavior outlier rule;
cluster a portion of the first data into a first cluster based on a statistical measure;
apply the behavior outlier rule to the first cluster to identify a first outlier and a second outlier;
generate a first alert for the first outlier and a second alert for the second outlier;
receive an indication of one or more user actions taken with respect to at least one of the first alert or the second alert; and
modify the behavior outlier rule based on the received indication using machine learning.

### Claim 3 (depends on 2)

The computing system of claim 2, wherein the program instructions further cause the computing system to validate the first cluster using historical transaction data.

### Claim 4 (depends on 2)

The computing system of claim 2, wherein the program instructions further cause the computing system to cluster a second portion of the first data into a second cluster.

### Claim 5 (depends on 4)

The computing system of claim 4, wherein the program instructions further cause the computing system to limit a number of clusters created from the first data using historical transaction data.

### Claim 6 (depends on 2)

The computing system of claim 2, wherein the first outlier and the second outlier are the top two outliers in the first cluster.

### Claim 7 (depends on 2)

The computing system of claim 2, wherein the one or more user actions comprises closing the first alert.

### Claim 8 (depends on 2)

The computing system of claim 2, wherein the program instructions further cause the computing system to:
identify a number of outliers including the first outlier and the second outlier that are actioned a threshold percentage of the time;
determine that a number of the one or more user actions taken with respect to the number of outliers including the first outlier and the second outlier is different than the threshold percentage; and
modify the behavior outliner rule in response to the determination that the number of the one or more user actions taken with respect to the number of outliers including the first outlier and the second outlier is different than the threshold percentage.

### Claim 9 (depends on 2)

The computing system of claim 2, wherein the statistical measure comprises one of a mean, a mode, or a standard deviation.

### Claim 10

A computer-implemented method comprising:
as implemented by one or more computer systems comprising computer hardware and memory, the one or more computer systems configured with specific executable instructions,
selecting a behavior outlier rule;
clustering a portion of first data into a first cluster based on a statistical measure;
applying the behavior outlier rule to the first cluster to identify a first outlier and a second outlier;
generating a first alert for the first outlier and a second alert for the second outlier;
receiving an indication of one or more user actions taken with respect to at least one of the first alert or the second alert; and
modifying the behavior outlier rule based on the received indication using machine learning.

### Claim 11 (depends on 10)

The computer-implemented method of claim 10, further comprising validating the first cluster using historical transaction data.

### Claim 12 (depends on 10)

The computer-implemented method of claim 10, further comprising clustering a second portion of the first data into a second cluster.

### Claim 13 (depends on 12)

The computer-implemented method of claim 12, further comprising limiting a number of clusters created from the first data using historical transaction data.

### Claim 14 (depends on 10)

The computer-implemented method of claim 10, wherein the first outlier and the second outlier are the top two outliers in the first cluster.

### Claim 15 (depends on 10)

The computer-implemented method of claim 10, wherein the one or more user actions comprises closing the first alert.

### Claim 16 (depends on 10)

The computer-implemented method of claim 10, wherein modifying the behavior outlier rule further comprises:
identifying a number of outliers including the first outlier and the second outlier that are actioned a threshold percentage of the time;
determining that a number of the one or more user actions taken with respect to the number of outliers including the first outlier and the second outlier is different than the threshold percentage; and
modifying the behavior outliner rule in response to the determination that the number of the one or more user actions taken with respect to the number of outliers including the first outlier and the second outlier is different than the threshold percentage.

### Claim 17 (depends on 10)

The computer-implemented method of claim 10, wherein the statistical measure comprises one of a mean, a mode, or a standard deviation.

### Claim 18

A non-transitory computer-readable medium comprising one or more program instructions recorded thereon, the instructions configured for execution by a computing system comprising one or more processors in order to cause the computing system to:
select a behavior outlier rule;
cluster a portion of first data into a first cluster based on a statistical measure;
apply the behavior outlier rule to the first cluster to identify a first outlier and a second outlier;
generate a first alert for the first outlier and a second alert for the second outlier;
receive an indication of one or more user actions taken with respect to at least one of the first alert or the second alert; and
modify the behavior outlier rule based on the received indication using machine learning.

### Claim 19 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein the instructions are further configured to cause the computing system to validate the first cluster using historical transaction data.

### Claim 20 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein the instructions are further configured to cause the computing system to:
identify a number of outliers including the first outlier and the second outlier that are actioned a threshold percentage of the time;
determine that a number of the one or more user actions taken with respect to the number of outliers including the first outlier and the second outlier is different than the threshold percentage; and
modify the behavior outliner rule in response to the determination that the number of the one or more user actions taken with respect to the number of outliers including the first outlier and the second outlier is different than the threshold percentage.

### Claim 21 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein the one or more user actions comprises closing the first alert.
