Systems and methods for measuring and improving literacy through deep learning and large-scale language model-assisted reading comprehension

Abstract

The methods and systems described herein provide a novel and secure approach to improving and measuring comprehension of technical, scientific, or other documents. By identifying esoteric concepts and providing additional secure, verified resources to aid understanding, the system enables readers to gain a deeper understanding of complex technical documents. The system can be implemented as a web-based application or as a plug-in to existing reading applications and can be customized for specific areas. The system works with existing documents; content owners do not need to recreate their content to fit a new format. The system and methods have broad availability in addition to making it easier for students to understand articles. The system provides feedback to content creators and enables teachers to identify students' efforts in reading documents. The system allows businesses to better engage with customers who are reading existing marketing and technical content online.

Claims

1. A computer system having at least one processor and a display to display user-readable text, In response to a user's selection of a word or phrase, the system provides generative artificial intelligence (AI) information relating to the selected word or phrase. A method for providing this.

2. The method according to claim 1, wherein the selection can be made by one of voice commands, taps, touches, gaze, or the use of a computer mouse.

3. The method according to claim 1, wherein the generated AI information is provided via a popup text box.

4. The method according to claim 3, wherein the popup text box includes a hyperlink to a website on the World Wide Web.

5. The method according to claim 3, wherein the popup text box is manually editable.

6. The method according to claim 1, wherein the generated AI information is provided via an audio message, a video message, or an image.

7. The method according to claim 1, further comprising providing different generated AI information when it is determined that the generated AI information is inaccurate or inappropriate.

8. The system further comprises providing, in response to a second selection of the word or phrase by the user, different generating AI information relating to the selected word or phrase from previously provided generating AI information, The aforementioned different information is generated by a machine learning model (MLL). The aforementioned MLL is trained using the captured data relating to the user, The method according to claim 1, wherein the different information is generated in real time by the MLL in response to the user's second selection of the word or phrase.

9. A non-temporary machine-readable medium having executable instructions that facilitate the execution of an operation when executed by at least one processor of the system, The aforementioned operation involves displaying user-readable text using a computer system comprising at least one processor and a display. In response to a user's selection of a word or phrase, the system provides generative artificial intelligence (AI) information relating to the selected word or phrase. A non-temporary machine-readable medium equipped with [a specific feature / feature].

10. The non-temporary machine-readable medium according to claim 9, wherein the selection can be made by one of voice commands, taps, touches, gaze, or the use of a computer mouse.

11. The generated AI information is provided via a pop-up text box in the non-temporary machine-readable medium according to claim 9.

12. The non-temporary machine-readable medium according to claim 11, wherein the popup text box comprises a hyperlink to a website on the World Wide Web.

13. The non-temporary machine-readable medium according to claim 11, wherein the pop-up text box is manually editable.

14. The non-temporary machine-readable medium according to claim 9, wherein the generated AI information is provided via an audio message, a video message, or an image.

15. The non-temporary machine-readable medium according to claim 9, further comprising providing different generated AI information when it is determined that the generated AI information is inaccurate or inappropriate.

16. The system further comprises providing, in response to a second selection of the word or phrase by the user, different generating AI information relating to the selected word or phrase from previously provided generating AI information, The aforementioned different information is generated by a machine learning model (MLL). The aforementioned MLL is trained using the captured data relating to the user, The non-temporary machine-readable medium according to claim 9, wherein the different information is generated in real time by the MLL in response to the user's second selection of the word or phrase.

17. A computer system having at least one processor and a display, A first component configured to display user-readable text on the aforementioned display, A second component configured to provide generative artificial intelligence (AI) information relating to a word or phrase selected by a user, in response to the user's selection of the word or phrase. A system equipped with these features.

18. The system according to claim 17, wherein the selection can be made by one of voice commands, taps, touches, gaze, or the use of a computer mouse.

19. The system according to claim 17, wherein the generated AI information is provided via a pop-up text box.

20. The system according to claim 17, wherein the popup text box is manually editable.

Images