# US 10,311,741 B2 · Data extraction and analysis system and tool

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

- Number: US 10,311,741 B2
- Document ID: US10311741B2
- Status: Granted
- Kind: B2
- Application number: 14/791,167
- Filed: 2015-07-02
- Published: 2017-01-05
- Granted: 2019-06-04
- Earliest claimed: 2015-07-02
- Assignees: Pearson Education, Inc.
- Applicants: Pearson Education, Inc.
- Inventors: Antonio Moretti; Kathy McKnight; José Pablo González Brenes
- Web page: https://patentgrep.com/patent/US10311741B2

## CPC

- G09B 5/00: Electrically-operated educational appliances (working with questions and answers G09B7/00; simulators G09B9/00; advertising or displaying in general G09F)
- G06Q 50/20: Education
- G09B 7/00: Electrically-operated teaching apparatus or devices working with questions and answers (mechanically operated G09B3/00; computing arrangements G06F)
- G09B: EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS (devices for psychotechnics or for testing reaction times A61B5/16; games, sports, amusements A63; projectors, projector screens G03B)
- 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

A data extraction and analysis system and tool is disclosed herein. The data extraction and analysis system and tool can include memory containing a comparison database, a factor database, and a model database that can include a multilevel model. The data extraction and analysis system and tool can include a content management server. The content management server can receive a request identifying a species and a variable and can retrieve data to generate a statistical model. Based on the statistical model, the content management server can identify and recommend an option to the requestor.

## Claims

### Claim 1

A data extraction system comprising:
a database server comprising:
a comparison database comprising data ranking options based on a comparison of statistical models;
a factor database; and
a model database, wherein the model database comprises a multilevel model, wherein the multilevel model comprises a statistical model of parameters that vary at more than one level; and
a content management server including:
a processor, and
a non-transitory memory storing instructions that when executed by the processor cause the processor to:
receive a request for a recommendation from a user device, wherein the request for a recommendation comprises data identifying a category and a variable;
identify a plurality of species data sources, wherein the species data sources comprise at least a first website hosted by a first content server hosting a first course content and a second website hosted by a second content server hosting a second course content;
retrieve species content from the species data sources, wherein the species content includes the first course content and the second course content;
parse the first course content to identify a first title of the first course content, a course associated with the first course content, and an education institution associated with the first course content;
parse the second course content to identify a second title of the second course content, a course associated with the second course content, and an education institution associated with the second course content;
identify, using the first title and the second title, a plurality of potential evaluation data sources, wherein the potential evaluation data sources comprise at least a third website hosted on a third content server and the third website hosts a first evaluation data including the first title and a second evaluation data including the second title;
retrieve potential evaluation data from the potential evaluation data sources, the potential evaluation data including the first evaluation data and the second evaluation data;
extract evaluation data from the retrieved potential evaluation data;
generate the multilevel model based on the species content and the evaluation data, by:
identifying a first subset of the species content, wherein all species content in the first subset is associated with the same course,
generating a first level simplified model at a first level of the species content and the evaluation data, wherein the first level simplified model is derived from the first subset of the species content,
identifying a second subset of the species content, wherein all species content in the second subset is associated with the same education institution,
generating a second level simplified model at a second level of the species content and the evaluation data, wherein the second level simplified model is derived from the second subset of the species content, and
estimating level variances for the first level simplified model and the second level simplified model, wherein the level variances comprise the percentage of variation in an aggregate of the species content and the evaluation data attributable to each of the first level simplified model and the second level simplified model; and
provide a recommendation of the first subset of the species content or a second subset of the species content by:
selecting, using the level variances, the first level simplified model or the second level simplified model, and
executing the selected one of the first level simplified model or the second level simplified model.

### Claim 2 (depends on 1)

The data extraction system of claim 1, wherein retrieving the species content comprises:
receiving format information, wherein the format information identifies the format in which the species content is contained in the species data source; and
identifying at least one relevant field in the species data of the species data source.

### Claim 3 (depends on 2)

The data extraction system of claim 2, wherein retrieving the species data further comprises scraping the species content from the at least one identified relevant field in the species data.

### Claim 4 (depends on 3)

The data extraction system of claim 3, wherein scraping the species content comprises web scraping.

### Claim 5 (depends on 4)

The data extraction system of claim 4, wherein retrieving the potential evaluation data comprises:
receiving format information, wherein the format information identifies the format in which the potential evaluation data is contained in the potential evaluation data source;
identifying at least one relevant field of the potential evaluation data source; and
scraping the potential evaluation data from the at least one identified relevant field of the potential evaluation data source.

### Claim 6 (depends on 1)

The data extraction system of claim 1, wherein extracting evaluation data information from the retrieved potential evaluation data comprises:
identifying ranking variables, wherein the ranking variables are stored within the model database;
identifying the evaluation data;
generating ranking data corresponding to the ranking variables based on the evaluation data; and
storing the ranking data.

### Claim 7 (depends on 1)

The data extraction system of claim 1, wherein the level variances are calculated as one of a variance partition coefficient (VPC) and an intraclass correlation coefficient (ICC).

### Claim 8 (depends on 7)

The data extraction system of claim 7, wherein the first level simplified model and the second level simplified model is selected according to at least one of a Chi-squared likelihood ratio test, an Akaike information criterion (AIC), and a Bayesian information criterion (BIC).

### Claim 9

A method of data extraction and analysis comprising:
receiving a request for a recommendation from a user device, wherein the request for a recommendation comprises data identifying a category and a variable;
identifying a plurality of species data sources, wherein the species data sources comprise at least a first website hosted by a first content server hosting a first course content and a second website hosted by a second content server hosting a second course content, and wherein the identified plurality of species data sources have a category matching the category of the received request;
retrieving species content from the species data sources, wherein the species content includes the first course content and the second course content;
parsing the first course content to identify a first title of the first course content, a course associated with the first course content, and an education institution associated with the first course content;
parsing the second course content to identify a second title of the second course content, a course associated with the second course content, and an education institution associated with the second course content;
identifying, using the first title and the second title, a plurality of potential evaluation data sources, wherein the potential evaluation data sources comprise at least a third website hosted on a third content server and the third website hosts a first evaluation data including the first title and a second evaluation data including the second title;
retrieving potential evaluation data from the potential evaluation data sources, the potential evaluation data including the first evaluation data and the second evaluation data;
extracting evaluation data from the retrieved potential evaluation data;
generating a multilevel model based on the species content and the evaluation data;
generating the multilevel model based on the species content and the potential evaluation data, by:
identifying a first subset of the species content, wherein all species content in the first subset is associated with the same course,
generating a first level simplified model at a first level of the species content and the evaluation data, wherein the first level simplified model is derived from the first subset of the species content,
identifying a second subset of the species content, wherein all species content in the second subset is associated with the same education institution,
generating a second level simplified model at a second level of the species content and the evaluation data, wherein the second level simplified model is derived from the second subset of the species content, and
estimating level variances for the first level simplified model and the second level simplified model, wherein the level variances comprise the percentage of variation in an aggregate of the species content and the evaluation data attributable to each of the first level simplified model and the second level simplified model; and
providing a recommendation based on the multilevel model by:
selecting, using the level variances, the first level simplified model or the second level simplified model, and
executing the selected one of the first level simplified model or the second level simplified model.

### Claim 10 (depends on 9)

The method of claim 9, wherein retrieving the species content comprises:
receiving format information, wherein the format information identifies the format in which the species content is contained in the species data source;
identifying at least one relevant field in the species data of the species data source; and scraping the species content from the at least one identified relevant field in the species data.

### Claim 11 (depends on 10)

The method of claim 10, wherein retrieving the potential evaluation data comprises:
receiving format information, wherein the format information identifies the format in which the potential evaluation data is contained in the potential evaluation data source;
identifying at least one relevant field of the potential evaluation data source; and
scraping the potential evaluation data from the at least one identified relevant field of the potential evaluation data source.

### Claim 12 (depends on 9)

The method of claim 9, wherein extracting evaluation data information from the retrieved potential evaluation data comprises:
identifying ranking variables, wherein the ranking variables are stored within a model database;
identifying the evaluation data;
generating ranking data corresponding to the ranking variables based on the evaluation data; and
storing the ranking data.

### Claim 13 (depends on 9)

The method of claim 9, wherein the level variances are calculated as one of a variance partition coefficient (VPC) and an intraclass correlation coefficient (ICC), and wherein the most accurate of the simplified model and the second simplified model is selected according to at least one of a Chi-squared likelihood ratio test, an Akaike information criterion (AIC), and a Bayesian information criterion (BIC).