# US 2016/0261618 A1 · System and method for selectively evolving phishing detection rules

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

- Number: US 2016/0261618 A1
- Document ID: US20160261618A1
- Status: Published application
- Kind: A1
- Application number: 15/012,418
- Filed: 2016-02-01
- Published: 2016-09-08
- Earliest claimed: 2015-03-05
- Assignees: AO Kaspersky Lab
- Inventors: Maxim G. Koshelev
- Web page: https://patentgrep.com/patent/US20160261618A1

## CPC

- H04L 63/1416: {Event detection, e.g. attack signature detection}
- G06F 16/285: {Clustering or classification}
- G06F 21/55: Detecting local intrusion or implementing counter-measures
- G06F 21/566: {Dynamic detection, i.e. detection performed at run-time, e.g. emulation, suspicious activities}
- H04L 51/212: using filtering or selective blocking
- H04L 63/0227: {Filtering policies (mail message filtering H04L51/212)}
- H04L 63/0236: {Filtering by address, protocol, port number or service, e.g. IP-address or URL}
- H04L 63/0263: {Rule management}
- H04L 63/14: {for detecting or protecting against malicious traffic}
- H04L 63/1433: {Vulnerability analysis}
- H04L 63/1441: {Countermeasures against malicious traffic (countermeasures against attacks on cryptographic mechanisms H04L9/002)}
- H04L 63/1483: {service impersonation, e.g. phishing, pharming or web spoofing (detection of rogue wireless access points H04W12/12)}
- H04L 63/20: {for managing network security; network security policies in general (filtering policies H04L63/0227)}
- H04L 63/0245: {Filtering by information in the payload}
- H04L 63/0281: {Proxies}
- 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)

## Abstract

System and method for automatically developing phishing detection rules. Based on detected phishing indicia, a quantitative score is computed for each of a plurality of predefined parameters, with each of the parameters relating to at least one of the phishing indicia. A requirement for evolving a phishing detection rule is assessed, and a new phishing detection rule is generated based on selected parameter scores meeting the rule evolution criteria and on corresponding content of the phishing indicia relating to those selected parameter scores. New phishing detection rules are applied recursively to detect phishing indicia, and more new rules can be further evolved in recursive fashion.

## Claims

### Claim at position 1

1-20. (canceled)

### Claim 21

A method for selectively evolving phishing detection rules with a computing system, the method comprising:
intercepting data with the computing system, the intercepted data being sent from a server and intended for a client;
examining, with the computing system, the intercepted data against a rule set, the rule set comprising a plurality of rules related to potential phishing indicia of the intercepted data, the examining including—
determining, as a result of the examining, a categorization from the plurality of categories of the intercepted data,
determining, as a result of the examining, at least one categorization-specific phishing detection criteria of the intercepted data, and
determining, as a result of the examining, at least one parameter score, the at least one parameter score comprising a likelihood of a presence of phishing content in the intercepted data based on at least the categorization-specific phishing detection criteria;
comparing the at least one parameter score against at least one rule generation criteria;
generating, with the computing system, a new rule and adding the new rule to the rule set to establish a second rule set if the at least one parameter score satisfies the at least one rule generation criteria; and
recursively examining, with the computing system, the intercepted data against the second rule set.

### Claim 22 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, wherein the at least one parameter score is a weighting factor or a flag.

### Claim 23 (depends on 22)

The method for selectively evolving phishing detection rules of claim 22, wherein the intercepted data is a string and the at least one parameter score is a weighting factor indicating a proportional value of a probability of the presence of phishing content.

### Claim 24 (depends on 22)

The method for selectively evolving phishing detection rules of claim 22, wherein the intercepted data is a hyperlink and the at least one parameter score is a flag indicating a trusted or a non-trusted link.

### Claim 25 (depends on 22)

The method for selectively evolving phishing detection rules of claim 22, wherein the at least one parameter score is a weighting factor and the at least one parameter score is greater than the at least one rule generation criteria.

### Claim 26 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, wherein the at least one parameter score satisfies the at least one rule generation criteria, the method further comprising evaluating the intercepted data for phishing-related content against the second rule set.

### Claim 27 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, wherein the at least one parameter score does not satisfy the at least one rule generation criteria and wherein the rule set is equivalent to the second rule set, the method further comprising evaluating the intercepted data for phishing-related content against the rule set.

### Claim 28 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, further comprising using previously-obtained rules of the rule set to increase the number of logical links between phishing indicia.

### Claim 29 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, further comprising using previously-obtained rules of the rule set to decrease the number of logical links between phishing indicia.

### Claim 30 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, wherein determining a categorization from the plurality of categories of the intercepted data further comprises determining a second categorization from the plurality of categories of the content of the intercepted data.

### Claim 31 (depends on 30)

The method for selectively evolving phishing detection rules of claim 30, wherein determining at least one categorization-specific phishing detection criteria of the intercepted data further comprises determining a second phishing detection criteria specific to the second categorization.

### Claim 32 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, wherein determining the at least one parameter score is based on at least the second phishing detection criteria.

### Claim 33 (depends on 21)

The method for selectively evolving phishing detection rules of claim 21, wherein the at least one rule generation criteria is an upper threshold.

### Claim 34 (depends on 31)

The method for selectively evolving phishing detection rules of claim 31, wherein the at least one rule generation criteria is at least two lower thresholds.

### Claim 35

A computing system for selectively evolving phishing detection rules, the system comprising:
means for intercepting data sent from a server and intended for a client;
means for examining the intercepted data against a rule set, the rule set comprising a plurality of rules related to potential phishing indicia of the intercepted data, the examining including—
means for determining a categorization from the plurality of categories of the intercepted data,
means for determining at least one categorization-specific phishing detection criteria of the intercepted data,
means for determining at least one parameter score, the at least one parameter score comprising a likelihood of a presence of phishing content in the intercepted data based on at least the categorization-specific phishing detection criteria,
means for comparing the at least one parameter score against at least one rule generation criteria;
means for generating a new rule and adding the new rule to the rule set to establish a second rule set with the computing system if the at least one parameter score satisfies the at least one rule generation criteria; and
means for recursively examining the intercepted data against the second rule set.

### Claim 36

A method for evolving phishing detection rules over a plurality of generations for a plurality of categories with a computing system, the method comprising:
generating, with the computing system, a first phishing detection rule applicable to a first category of data based on a set of intercepted data;
applying, with the computing system, the first phishing detection rule to the set of intercepted data in accordance with the first category of data, the first category of data including a first set of phishing indicators specific to the first category and a first set of parameters specific to the first category;
evaluating, with the computing system, a decision criteria for evolving a subsequent generation of rules based on the first phishing detection rule application to the set of intercepted data;
if the decision criteria is satisfied, generating, with the computing system, a second phishing detection rule applicable to a second category of data, the second category of data being different than the first category of data; and
applying, with the computing system, the second phishing detection rule to the set of intercepted data in accordance with the second category of data, the second category of data including a second set of phishing indicators specific to the second category and a second set of parameters specific to the second category.

### Claim 37 (depends on 36)

The method for evolving phishing detection rules of claim 36, wherein the first set of parameters have a one-to-one correspondence, a many-to-one correspondence, or a one-to-many correspondence with the first set of phishing indicators.

### Claim 38 (depends on 36)

The method for evolving phishing detection rules of claim 36, further comprising:
evaluating, with the computing system, a second decision criteria for evolving a subsequent generation of rules based on the second phishing detection rule application to the set of intercepted data;
if the second decision criteria is satisfied, generating, with the computing system, a third phishing detection rule applicable to the first category of data; and
applying, with the computing system, the third phishing detection rule to the set of intercepted data in accordance with the first category of data.

### Claim 39 (depends on 36)

The method for evolving phishing detection rules of claim 36, further comprising:
evaluating, with the computing system, a second decision criteria for evolving a subsequent generation of rules based on the second phishing detection rule application to the set of intercepted data; and
if the second decision criteria is satisfied, generating, with the computing system, a third phishing detection rule,
wherein the second decision criteria is of a higher resolution than the first decision criteria.

### Claim 40 (depends on 36)

The method for evolving phishing detection rules of claim 36, wherein the first category of data is a software code object and the second category of data is an element within the software code object.
