Method and Apparatus for Performing Dynamic Textual Complexity Analysis Using Machine Learning Artificial Intelligence

• US 20130282628A1
• Filed: 08/22/2012
• Published: 10/24/2013
• Est. Priority Date: 04/20/2012
• Status: Active Grant

• Current Assignee: Certica Solutions Incorporated
• Original Assignee: Unbound Concepts LLC
• Inventors: Bengfort, Benjamin, Palencsar, Katie, Voorhees, William
• Granted Patent: US 8,880,445 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

A data processing system including one or more client devices, wherein each client device is connected to a network system and a data center unit. The data center unit includes a network interface unit, a user interface, one or more storage devices, wherein the one or more storage devices comprise one or more databases. Further, the data center unit includes a storage device controller and database manager for controlling the operation of storage devices and databases, a web server for providing web services to clients, a database server for providing database services to the one or more clients and a machine learning artificial intelligence application server for predicting textual complexity of data. The machine learning artificial intelligence application server includes one or more databases for storing data used to refine textual complexity analysis for improved accuracy of textual complexity predictions.

Claims

1. A method of determining textual complexity by performing dynamic analysis of text using machine learning artificial intelligence, comprising the steps of:

a) receiving, in an application server, input data;

b) formatting, by the application server, the received input data;

c) storing, by the application server, the formatted data in a storage device creating a digital library;

d) performing, by the application server, quantitative analysis of the input data in parallel with the application server performing qualitative analysis of said input data;

e) executing complexity analysis, by the application server, of the input data using data generated from said performed quantitative analysis and qualitative analysis to predict the textual complexity of the input data;

f) refining, by the application server, the complexity analysis of said input data using the machine learning artificial intelligence; and

g) providing improved accuracy of textual complexity predictions based on the refined complexity analysis.

2. The method of claim 1, wherein the machine learning artificial intelligence uses continuous feedback data to perform said refining of predictions.

3. The method of claim 2, wherein the continuous feedback data includes supervised training information and feature resolution data.

4. The method of claim 3, wherein the supervised training information is generated from a supervised training session.

5. The method of claim 4, wherein the supervised training session receives ranking and rating information from a user.

6. The method of claim 5, wherein an accuracy evaluation is made, by the application server, based on statistical analysis of said user ranking and rating information.

7. The method of claim 6, wherein the machine learning artificial intelligence uses the accuracy determination made by the application server to perform said refining of predictions.

8. The method of claim 3, wherein the feature resolution data is derived from a polynomial best fit line analysis.

9. The method of claim 8, wherein the polynomial best fit line analysis generates feature modification parameters.

10. The method of claim 8, wherein the feature modification parameters include feature id, range start, range end, and weight.

11. The method of claim 3, wherein steps d) and e) are repeated for each entry of data stored in the digital library when feature resolution data is received.

12. The method of claim 1, wherein the quantitative analysis determines structural complexity of data by evaluating features identified in the data.

13. The method of claim 12, wherein the features are evaluated using a weight factor.

14. The method of claim 13, wherein the weight factor establishes the importance of a feature relative to other features in the determination of textual complexity.

15. The method of claim 1, wherein the qualitative analysis determines the complexity of the concepts of the data.

16. A data processing system for performing dynamic textual complexity analysis using machine learning artificial intelligence comprising:

one or more client devices, wherein each client device is connected to a network system;

a data center unit comprising;

a network interface unit for interfacing with the one or more client devices and the network system;

a user interface;

one or more storage devices, wherein the one or more storage devices comprise one or more databases;

a storage device controller and database manager for controlling the operations of the one or more storage devices and one or more databases;

a web server for providing web services to the one or more clients; and

a database server for providing database services to the one or more clients;

a machine learning artificial intelligence (“

MLAI”

) application server for predicting textual complexity of data, the machine learning artificial intelligence application server includes one or more databases for storing data used to refine textual complexity analysis for improved accuracy of textual complexity predictions.

17. The data processing system of claim 16, wherein the one or more storage devices includes a digital library.

18. The data processing system of claim 16, wherein the MLAI application server includes a first database for storing supervised training information.

19. The data processing system of claim 16, wherein the MLAI application server includes a second database for storing feature resolution data.

20. The data processing system of claim 16, wherein the MLAI application server uses continuous feedback data to perform said refining of textual complexity analysis.