# US 2019/0171756 A1 · Cognitive decision system for security and log analysis using associative memory mapping in graph database

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

- Number: US 2019/0171756 A1
- Document ID: US20190171756A1
- Status: Published application
- Kind: A1
- Application number: 16/119,461
- Filed: 2018-08-31
- Published: 2019-06-06
- Earliest claimed: 2017-12-04
- Assignees: Accenture Global Solutions Limited
- Applicants: Accenture Global Solutions Limited
- Inventors: Sudhir Ranganna Patavardhan; Nikhil S. Tanwar
- Web page: https://patentgrep.com/patent/US20190171756A1

## CPC

- G06F 16/288: {Entity relationship models}
- G06F 16/2379: {Updates performed during online database operations; commit processing}
- G06F 16/3344: {using natural language analysis}
- G06F 16/367: {Ontology}
- G06F 16/38: Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F 16/9024: {Graphs; Linked lists (G06F16/9027 takes precedence)}
- G06F 21/6227: {where protection concerns the structure of data, e.g. records, types, queries}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a system to create and employ associative memory maps for analysis of security file and/or logs are disclosed. In one aspect, a method includes the actions of receiving, from an external application, a request for a recommended action; extracting information regarding the entities and relationships between the entities from a data source; constructing an associative memory map from the extracted information; selecting a subgraph from the associative memory map based on a result of employing a vector to search nodes in the associative memory map; identifying the nodes most relevant to the requested recommend action base on a shortest paths of traversal in the selected subgraph of nodes; determining the requested recommended action based on an event identified in the relationships between the identified most relevant nodes; and transmitting the recommended action to the external application.

## Claims

### Claim 1

A computer-implemented method executed by one or more processors, the method comprising:
receiving, from an external application, a request for a recommended action regarding entities associated with a source system;
extracting information regarding the entities and relationships between the entities from a data source in which the source system persists transaction and security data;
constructing an associative memory map from the extracted information, wherein the associative memory map describes the entities as nodes and characterizes a strength of the relationships between each entity;
selecting a subgraph of nodes from the associative memory map based on a result of employing a vector to search the nodes in the associative memory map, the vector generated based on the requested recommended action;
identifying the nodes most relevant to the requested recommend action base on a shortest paths of traversal in the selected subgraph of nodes;
determining the requested recommended action based on an event identified in the relationships between the identified most relevant nodes; and
transmitting the recommended action to the external application.

### Claim 2 (depends on 1)

The method of claim 1, wherein the associative memory map is constructed based on ontologies identified based on an analysis of metadata from the transaction and security data.

### Claim 3 (depends on 1)

The method of claim 1, wherein the data source comprises a relational database management system (RDBMS) and the transaction and security data is stored in a relational model.

### Claim 4 (depends on 3)

The method of claim 3, wherein extracting information regarding the entities and the relationships between the entities includes retailing the information from the RDBMS and employing a graph mappings using key constrains from columns in RDBMS.

### Claim 5 (depends on 1)

The method of claim 1, wherein the data source comprises structured files and the transaction and security data is stored in fixed fields within a record or file.

### Claim 6 (depends on 1)

The method of claim 1, wherein the data source comprises text and log files and the transaction and security data is stored as unstructured data.

### Claim 7 (depends on 6)

The method of claim 6, wherein extracting information regarding the entities and the relationships between the entities includes natural language processing (NLP) and regular expression analysis of the stored unstructured data.

### Claim 8 (depends on 1)

The method of claim 1, wherein the data source comprises audio files or video files, and wherein extracting information regarding the entities and the relationships between the entities includes an analysis of metadata of the audio files or the video files.

### Claim 9 (depends on 1)

The method of claim 1, wherein the event comprises a high threshold event, a rare relationship event, a rare node and relationship event, or an unusual pattern event.

### Claim 10 (depends on 1)

The method of claim 1, further comprising:
transmitting the event to the external application.

### Claim 11

One or more non-transitory computer-readable storage media coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, from an external application, a request for a recommended action regarding entities associated with a source system;
extracting information regarding the entities and relationships between the entities from a data source in which the source system persists transaction and security data;
constructing an associative memory map from the extracted information, wherein the associative memory map describes the entities as nodes and characterizes a strength of the relationships between each entity;
selecting a subgraph of nodes from the associative memory map based on a result of employing a vector to search the nodes in the associative memory map, the vector generated based on the requested recommended action;
identifying the nodes most relevant to the requested recommend action base on a shortest paths of traversal in the selected subgraph of nodes;
determining the requested recommended action based on an event identified in the relationships between the identified most relevant nodes; and
transmitting the recommended action to the external application.

### Claim 12 (depends on 11)

The one or more non-transitory computer-readable storage media of claim 11, wherein the associative memory map is constructed based on ontologies identified based on an analysis of metadata from the transaction and security data.

### Claim 13 (depends on 11)

The one or more non-transitory computer-readable storage media of claim 11, wherein the data source comprises a relational database management system (RDBMS) and the transaction and security data is stored in a relational model.

### Claim 14 (depends on 13)

The one or more non-transitory computer-readable storage media of claim 13, wherein extracting information regarding the entities and the relationships between the entities includes retailing the information from the RDBMS and employing a graph mappings using key constrains from columns in RDBMS.

### Claim 15 (depends on 11)

The one or more non-transitory computer-readable storage media of claim 11, wherein the data source comprises structured files and the transaction and security data is stored in fixed fields within a record or file.

### Claim 16

A system, comprising:
one or more processors; and
a computer-readable storage device coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, from an external application, a request for a recommended action regarding entities associated with a source system;
extracting information regarding the entities and relationships between the entities from a data source in which the source system persists transaction and security data;
constructing an associative memory map from the extracted information, wherein the associative memory map describes the entities as nodes and characterizes a strength of the relationships between each entity;
selecting a subgraph of nodes from the associative memory map based on a result of employing a vector to search the nodes in the associative memory map, the vector generated based on the requested recommended action;
identifying the nodes most relevant to the requested recommend action base on a shortest paths of traversal in the selected subgraph of nodes;
determining the requested recommended action based on an event identified in the relationships between the identified most relevant nodes; and
transmitting the recommended action to the external application.

### Claim 17 (depends on 16)

The system of claim 16, wherein the data source comprises audio files or video files, and wherein extracting information regarding the entities and the relationships between the entities includes an analysis of metadata of the audio files or the video files.

### Claim 18 (depends on 16)

The system of claim 16, wherein the transaction and security data are stored as records of captured sound or digital video that can be played back, wherein each record of captured sound or digital video is extracted from a respective audio or visual file.

### Claim 19 (depends on 18)

The system of claim 18, wherein the event comprises a high threshold event, a rare relationship event, a rare node and relationship event, or an unusual pattern event.

### Claim 20 (depends on 19)

The system of claim 19, the operations further comprising:
transmitting the event to the external application.
