# US 2018/0324213 A1 · Methods and systems for cyber-hacking detection

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

- Number: US 2018/0324213 A1
- Document ID: US20180324213A1
- Status: Published application
- Kind: A1
- Application number: 15/585,012
- Filed: 2017-05-02
- Published: 2018-11-08
- Earliest claimed: 2017-05-02
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Matthew G. Borlick; Lokesh M. Gupta
- Web page: https://patentgrep.com/patent/US20180324213A1

## CPC

- H04L 63/1491: {using deception as countermeasure, e.g. honeypots, honeynets, decoys or entrapment}
- G06F 21/552: {involving long-term monitoring or reporting}
- G06N 20/00: Machine learning
- G06Q 20/227: {characterised in that multiple accounts are available, e.g. to the payer}
- G06Q 20/4016: {involving fraud or risk level assessment in transaction processing}
- H04L 63/1416: {Event detection, e.g. attack signature detection}
- G06F 2221/2111: Location-sensitive, e.g. geographical location, GPS
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- 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

## Abstract

Methods and systems for cyber-hacking detection are provided. One method includes generating, by a processor, one or more artificial accounts for a type of actual account, learning one or more hacking behaviors for the type of actual account, and detecting cyber-hacks in activity in the one or more artificial accounts based on the one or more hacking behaviors. Systems and computer program products for performing the above method are also provided.

## Claims

### Claim 1

An apparatus, comprising:
a generation module for generating one or more first artificial accounts for a first type of account;
a learning module for learning one or more first hacking behaviors for the first type of account; and
a detection module for detecting first cyber-hacks in first activity in the one or more first artificial accounts based on the one or more first hacking behaviors.

### Claim 2 (depends on 1)

The apparatus of claim 1, wherein the generation module further groups the one or more first artificial accounts in a valid account store comprising a plurality of actual accounts for the first type of account.

### Claim 3 (depends on 2)

The apparatus of claim 2, wherein the one or more first artificial accounts comprise a portion of the valid account store in the range of one percent to twenty-five percent.

### Claim 4 (depends on 1)

The apparatus of claim 1, wherein:
the generation module further generates one or more second artificial accounts associated with one or more additional types of account;
each of the one or more additional type of account includes a respective second plurality of actual accounts;
the learning module further learns one or more second hacking behaviors for each second type of account; and
the detection module further detects second cyber-hacks in second activity in the one or more second artificial accounts based on the one or more second hacking behaviors.

### Claim 5 (depends on 4)

The apparatus of claim 4, wherein:
the generation module further groups the one or more first artificial accounts and the one or more second artificial accounts in a valid account store comprising a first plurality of actual accounts and a second plurality of actual accounts for the first type of account;
the first plurality of actual accounts are managed by a first entity;
the second plurality of actual accounts are managed by a second entity; and
the detection module further distinguishes between a general cyber-hack based on the first type of account and a specific cyber-hack based on one of the first entity and the second entity.

### Claim 6 (depends on 4)

The apparatus of claim 4, wherein:
the generation module further:
groups the one or more first artificial accounts in a first valid account store comprising a first plurality of actual accounts for the first type of account, and
groups the one or more second artificial accounts in a second valid account store comprising a second plurality of actual accounts for the second type of account;
the first type of account and the second type of account are different types of account;
the first plurality of actual accounts and the second plurality of actual accounts are managed by a same entity; and
the detection module further distinguishes between a general cyber-hack based on the entity and a specific cyber-hack based on one of the first type of account and the second type of account.

### Claim 7 (depends on 1)

The apparatus of claim 1, further comprising:
a monitoring module for monitoring the one or more first artificial accounts for the first activity, wherein:
the first type of account is a financial account,
the first activity is at least one of sign-in activity and transaction activity for the one or more first artificial accounts, and
the detection module detects a first cyber-hack in response to an artificial account experiencing an attempt to at least one of sign-in to the artificial account and initiate a transaction associated with the artificial account.

### Claim 8 (depends on 1)

The apparatus of claim 1, wherein the learning module learns the one or more first hacking behaviors by one of:
learning one or more first user behaviors based on first user activity for a plurality of actual accounts associated with the first type of account and determining that a deviation from the one or more user behaviors defines the one or more first hacking behaviors; and
learning the one or more first hacking behaviors from hacking activity associated with one or more hacked accounts.

### Claim 9 (depends on 1)

The apparatus of claim 1, further comprising:
a monitoring module for monitoring the one or more first artificial accounts for the first activity, wherein:
the first type of account is an email account,
the first activity is at least one of sign-in activity and mailbox activity for the one or more first artificial accounts,
the sign-in activity comprises at least one of one or more login dates, one or more real-world locations from which the one or more logins originated, and an internet protocol (IP) number for each device utilized for the one or more logins,
the mailbox activity comprises email activity in at least one of a mail sent folder, a mail received folder, and a trash folder,
for sign-in activity, the detection module determines a first cyber hack if the at least one of determined one or more login dates, the determined one or more real-world locations, and the determined IP number for each device does not match a corresponding at least one of one or more legitimate login dates, one or more real-world locations from which the one or more legitimate logins originated, and the IP number for each legitimate device utilized for the one or more actual logins, and
for mailbox activity, the detection module:
compares email activity in the at least one of the mail sent folder, the mail received folder, and the trash folder to known email activity in the at least one of the mail sent folder, the mail received folder, and the trash folder, and
determines the first cyber hack if the email activity does not match the known activity.

### Claim 10

A method, comprising:
generating, by a processor, one or more first artificial accounts for a first type of actual account;
learning one or more first hacking behaviors for the first type of actual account; and
detecting first cyber-hacks in first activity in the one or more first artificial accounts based on the one or more first hacking behaviors.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
grouping the one or more first artificial accounts in a valid account store comprising a plurality of actual accounts for the first type of account, wherein the one or more first artificial accounts comprise a portion of the valid account store in the range of one percent to twenty-five percent.

### Claim 12 (depends on 10)

The method of claim 10, further comprising:
generating, by the processor, one or more second artificial accounts associated with one or more additional types of account, wherein each of the one or more additional type of account includes a respective second plurality of actual accounts;
learning one or more second hacking behaviors for each additional type of account; and
detecting second cyber-hacks in second activity in the one or more second artificial accounts based on the one or more second hacking behaviors.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
grouping the one or more first artificial accounts and the one or more second artificial accounts in a valid account store comprising a first plurality of actual accounts and a second plurality of actual accounts for the first type of account, wherein the first plurality of actual accounts are managed by a first entity and the second plurality of actual accounts are managed by a second entity; and
distinguishing between a general cyber-hack based on the first type of account and a specific cyber-hack based on one of the first entity and the second entity.

### Claim 14 (depends on 12)

The method of claim 12, further comprising:
grouping the one or more first artificial accounts in a first valid account store comprising a first plurality of actual accounts for the first type of account;
grouping the one or more second artificial accounts in a second valid account store comprising a second plurality of actual accounts for an additional type of account, wherein the first type of account and the additional type of account are different types of account, and the first plurality of actual accounts and the second plurality of actual accounts are managed by a same entity; and
distinguishing between a general cyber-hack based on the entity and a specific cyber-hack based on one of the first type of account and the additional type of account.

### Claim 15 (depends on 10)

The method of claim 10, wherein learning the one or more first hacking behaviors comprises one of:
learning one or more first user behaviors based on user activity for a plurality of actual accounts associated with the first type of account, wherein a deviation from the one or more user behaviors defines the one or more first hacking behaviors; and
learning the one or more first hacking behaviors from hacking activity associated with one or more hacked accounts.

### Claim 16

A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
generate one or more first artificial accounts for a first type of actual account;
learn one or more first hacking behaviors for the first type of actual account; and
detect first cyber-hacks in first activity in the one or more first artificial accounts based on the one or more first hacking behaviors.

### Claim 17 (depends on 16)

The computer program product of claim 16, wherein the program instructions further cause the processor to:
group the one or more first artificial accounts in a valid account store comprising a plurality of actual accounts for the first type of account, wherein the one or more first artificial accounts comprise a portion of the valid account store in the range of one percent to twenty-five percent.

### Claim 18 (depends on 17)

The computer program product of claim 17, wherein the program instructions further cause the processor to:
generate one or more second artificial accounts associated with one or more additional types of account, wherein each of the one or more additional type of account includes a respective second plurality of actual accounts;
learn one or more second hacking behaviors for each additional type of account; and
detect second cyber-hacks in second activity in the one or more second artificial accounts based on the one or more second hacking behaviors.

### Claim 19 (depends on 17)

The computer program product of claim 17, wherein the program instructions further cause the processor to:
group the one or more first artificial accounts in a first valid account store comprising a first plurality of actual accounts for the first type of account;
group the one or more second artificial accounts in a second valid account store comprising a second plurality of actual accounts for an additional type of account, wherein the first type of account and the additional type of account are different types of account, and the first plurality of actual accounts and the second plurality of actual accounts are managed by a same entity; and
distinguish between a general cyber-hack based on the entity and a specific cyber-hack based on one of the first type of account and the additional type of account.

### Claim 20 (depends on 16)

The computer program product of claim 16, wherein the program instructions causing the processor to learn the one or more first hacking behaviors further cause the processor to one of:
learn one or more user behaviors based on user activity for a plurality of actual accounts associated with the first type of account, wherein a deviation from the one or more user behaviors defines the one or more first hacking behaviors; and
learn the one or more first hacking behaviors from hacking activity associated with one or more hacked accounts.
