# US 2021/0342343 A1 · Data harvester

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

- Number: US 2021/0342343 A1
- Document ID: US20210342343A1
- Status: Published application
- Kind: A1
- Application number: 16/865,662
- Filed: 2020-05-04
- Published: 2021-11-04
- Earliest claimed: 2020-05-04
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Ashwin Kumar; Ankur Bharatkumar Shah; Atul Gore; Jayanth Gangadhar; Vivek Venkatanarasaiah
- Web page: https://patentgrep.com/patent/US20210342343A1

## CPC

- G06F 16/2465: {Query processing support for facilitating data mining operations in structured databases}
- G06F 16/9038: Presentation of query results
- G06N 5/02: Knowledge representation; Symbolic representation
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A data harvester enhances compliance audits by characterizing data sources, sampling data in one or more of the data sources to determine likelihood of success of the data harvest, estimating time for the data harvest, making recommendations from the samples based on machine learning relating to previous runs, then sampling additional data while estimated expected completion time. The harvested data may then be analyzed and compared to compliance requirements, and a compliance report may be generated.

## Claims

### Claim 1

An apparatus comprising:
at least one processor;
a memory coupled to the at least one processor; and
a data harvester residing in the memory and executed by the at least one processor, wherein the data harvester characterizes a plurality of data sources in a target system, samples a subset of data in the plurality of data sources, and estimates a time of completion based on the samples.

### Claim 2 (depends on 1)

The apparatus of claim 1 wherein the data harvester characterizes a selected data source according to a plurality of the following:
total size of the selected data source;
total number of documents in the selected data source;
types of documents in the selected data source; and
network characteristics of the selected data source.

### Claim 3 (depends on 2)

The apparatus of claim 2 wherein, when the selected data source includes a plurality of mailboxes, the data harvester further characterizes the selected data source according to organization, mailbox sizes and attachment percentages.

### Claim 4 (depends on 1)

The apparatus of claim 1 wherein the data harvester determines from the samples a likelihood of success of harvesting data, and when the likelihood of success is below a threshold, the data harvester selects a new sample size based on adjusted parameters for the data harvester.

### Claim 5 (depends on 1)

The apparatus of claim 1 wherein the data harvester generates based on machine learning at least one recommendation based on historical data stored in a knowledgebase regarding at least one previous run of the data harvester.

### Claim 6 (depends on 5)

The apparatus of claim 5 wherein a user selects at least one adjustment based on the at last one recommendation, and in response, the data harvester continues harvesting data according to the user-selected at least one adjustment.

### Claim 7 (depends on 6)

The apparatus of claim 6 wherein the data harvester selects a next sample, estimates time for processing the selected next sample, and performs data harvesting on the selected next sample.

### Claim 8

An article of manufacture comprising software stored on a computer readable storage medium, the software comprising:
a data harvester that characterizes a plurality of data sources in a target system, samples a subset of data in the plurality of data sources, and estimates a time of completion based on the samples.

### Claim 9 (depends on 8)

The article of manufacture of claim 8 wherein the data harvester characterizes a selected data source according to a plurality of the following:
total size of the selected data source;
total number of documents in the selected data source;
types of documents in the selected data source;
network characteristics of the selected data source; and
when the selected data source includes a plurality of mailboxes, according to organization, mailbox sizes and attachment percentages.

### Claim 10 (depends on 8)

The article of manufacture of claim 8 wherein the data harvester determines from the samples a likelihood of success of harvesting data, and when the likelihood of success is below a threshold, the data harvester selects a new sample size based on adjusted parameters for the data harvester.

### Claim 11 (depends on 8)

The article of manufacture of claim 8 wherein the data harvester generates based on machine learning at least one recommendation based on historical data stored in a knowledgebase regarding at least one previous run of the data harvester.

### Claim 12 (depends on 11)

The article of manufacture of claim 11 wherein a user selects at least one adjustment based on the at last one recommendation, and in response, the data harvester continues harvesting data according to the user-selected at least one adjustment.

### Claim 13 (depends on 12)

The article of manufacture of claim 12 wherein the data harvester selects a next sample, estimates time for processing the selected next sample, and performs data harvesting on the selected next sample.

### Claim 14

A method for harvesting data, the method comprising:
characterizing a plurality of data sources in a target system;
sampling a subset of data in the plurality of data sources; and
estimating a time of completion based on the samples.

### Claim 15 (depends on 14)

The method of claim 14 wherein characterizing a plurality of data sources in the target system comprises characterizing a selected data source according to a plurality of the following:
total size of the selected data source;
total number of documents in the selected data source;
types of documents in the selected data source; and
network characteristics of the selected data source.

### Claim 16 (depends on 15)

The method of claim 15 wherein, when the selected data source includes a plurality of mailboxes, characterizing the selected data source according to organization, mailbox sizes and attachment percentages.

### Claim 17 (depends on 14)

The method of claim 14 further comprising:
determining from the samples a likelihood of success of harvesting data; and
when the likelihood of success is below a threshold, selecting a new sample size based on adjusted parameters for the data harvester.

### Claim 18 (depends on 14)

The method of claim 14 further comprising generating based on machine learning at least one recommendation based on historical data stored in a knowledgebase regarding at least one previous run of the data harvester.

### Claim 19 (depends on 18)

The method of claim 18 wherein a user selects at least one adjustment based on the at last one recommendation, and in response, continuing harvesting data according to the user-selected at least one adjustment.

### Claim 20 (depends on 19)

The method of claim 19 further comprising:
selecting a next sample;
estimating time for processing the selected next sample; and
performing data harvesting on the selected next sample.
