# US 2019/0244146 A1 · Elastic distribution queuing of mass data for the use in director driven company assessment

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

- Number: US 2019/0244146 A1
- Document ID: US20190244146A1
- Status: Published application
- Kind: A1
- Application number: 16/250,744
- Filed: 2019-01-17
- Published: 2019-08-08
- Earliest claimed: 2018-01-18
- Assignees: D&B Business Information Solutions
- Applicants: D&B Business Information Solutions
- Inventors: Eoin Lane; Ciara Keady; Maria McGourty; Joke O'Connor
- Web page: https://patentgrep.com/patent/US20190244146A1

## CPC

- G06Q 10/0635: Risk analysis of enterprise or organisation activities
- G06F 16/24568: {Data stream processing; Continuous queries}
- G06F 16/254: {Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses}
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- G06N 20/00: Machine learning
- G06Q 30/0282: Rating or review of business operators or products
- 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)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

An elastic distribution queuing system for mass data comprising: a data source; a matching engine for matching and/or appending a corporate identifier to data from the data source, thereby creating enhanced data; a distributed queuing system which determines how much the enhanced data is being ingested by the distributed queuing system and how many distributed processing nodes will be required to process the enhanced data; a structured streaming engine for distributed processing of the enhanced data from each the distributed processing node; a decision tree engine which identifies at least one data element from the enhanced data and determines a value of importance of the data element; a logistic regression model which determines the probability of failure of a corporate entity associated with the enhanced data based upon the value of importance of the data element; and an output of the results from the logistic regression model regarding the probability of failure for the corporate entity.

## Claims

### Claim 1

An elastic distribution queuing system for mass data comprising:
a data source;
a matching engine for matching and/or appending a corporate identifier to data from said data source, thereby creating enhanced data;
a distributed queuing system which determines how much said enhanced data is being ingested by said distributed queuing system and how many distributed processing nodes will be required to process said enhanced data;
a structured streaming engine for distributed processing of said enhanced data from each said distributed processing node;
a decision tree engine which identifies at least one data element from said enhanced data and determines a value of importance of said data element;
a logistic regression model which determines the probability of failure of a corporate entity associated with said enhanced data based upon said value of importance of said data element; and
an output of the results from said logistic regression model regarding said probability of failure for said corporate entity.

### Claim 2 (depends on 1)

The system according to claim 1, wherein said distributed queuing system is a grate extract, transform and load queuing system.

### Claim 3 (depends on 1)

The system according to claim 1, wherein said distributed processing node is an elastic scalable distributed queueing system which processes said enhanced data in near real time across said structured streaming engine.

### Claim 4 (depends on 3)

The system according to claim 3, wherein the output further comprises a real-time alert to a downstream application.

### Claim 5 (depends on 1)

The system according to claim 1, wherein said structured streaming engine comprises at least one Spark node and a Spark engine.

### Claim 6 (depends on 5)

The system according to claim 5, wherein said spark engine enables incremental updates to be appended to said enhanced data.

### Claim 7 (depends on 1)

The system according to claim 1, further comprising machine learning by (a) learning the data element in the decision tree engine to confirm a feature set, and (b) said logistic regression model uses said feature set to train or test a data set to predict, thereby producing said probability of failure for said corporate entity.

### Claim 8 (depends on 3)

The system according to claim 3, wherein said elastic scalable distributed queueing system is a Kafka node.

### Claim 9

A method for elastic distribution queuing of mass data, the method being performed by a computer system that comprises distributed processors, a memory operatively coupled to at least one of the distributed processors, and a computer-readable storage medium encoded with instructions executable by at least one of the distributed processors and operatively coupled to at least one of the distributed processors, the method comprising:
retrieving data from at least one data source;
matching and/or appending a corporate identifier to saud data from said data source, thereby creating enhanced data;
distributed queuing of said enhanced data to determine how much of said enhanced data is being created and how many distributed processing nodes will be activated to process said enhanced data;
distributed processing of said enhanced data from each said distributed processing node via a structured streaming engine;
identifying at least one data element from said enhanced data and determining a value of importance of said data element via a decision tree engine;
determining the probability of failure of a corporate entity associated with said enhanced data based upon said value of importance of said data element via a logistic regression model; and
outputting of the results from said logistic regression model regarding said probability of failure for said corporate entity.

### Claim 10 (depends on 9)

The method according to claim 9, wherein said distributed queuing is performed by a grate extract, transform and load queuing system.

### Claim 11 (depends on 9)

The method according to claim 9, wherein said distributed processing node is an elastic scalable distributed queueing system which processes said enhanced data in near real time across said structured streaming engine.

### Claim 12 (depends on 11)

The method of claim 11, further comprising: outputting a real-time alert to a downstream application.

### Claim 13 (depends on 9)

The method according to claim 9, wherein said structured streaming engine comprises at least one Spark node and a Spark engine.

### Claim 14 (depends on 13)

The method according to claim 13, wherein said Spark engine enables incremental updates to be appended to said enhanced data.

### Claim 15 (depends on 9)

The method according to claim 9, further comprising (a) learning the data element in the decision tree engine to confirm a feature set, and (b) said logistic regression model uses said feature set to train or test a data set to predict, thereby producing said probability of failure for said corporate entity

### Claim 16 (depends on 11)

The method according to claim 11, wherein said elastic scalable distributed queueing system is a Kafka nod
