Information and dialog models for education
A reading comprehension system may include an authoring tool to help generate adaptable dialogs and a reading tool to conduct adaptable dialog sessions with students. The authoring tool may receive and process stories to generate labeled stories and information models. The information models may provide the conceptual structures that an effective reader should build while reading and understanding a story. The system may use dialog models for general dialogs and information models for story specific dialogs to guide adaptable dialog sessions with students. During the adaptable dialog sessions, the system may constantly assess and guide the student's progress in the understanding the current story and in general reading comprehension development. Using the labeled stories and dialog sessions as training data, the system may learn how to dialog effectively with the students, to gather an evolving understanding of the student's abilities, and to acquire knowledge about the world or the story.
1. A computer-implemented method comprising:
receiving, from a first device associated with a first user, a story, wherein the first user is associated with an author account for a reading comprehension tool;
processing, by the reading comprehension tool, the story to identify one or more elements in the story, the one or more elements including at least one of a character, an event, or a setting associated with the story;
generating, by an authoring tool component associated with the reading comprehension tool, a tagged story model by tagging individual elements of the one or more elements with one or more labels;
generating, by the authoring tool component and based at least in part on the tagged story model, an adaptable dialog set associated with the one or more elements, wherein the adaptable dialog set includes one or more questions for evaluating reading comprehension level;
determining, based at least in part on the one or more labels in the tagged story model, at least one answer for the one or more questions;
generating an information model of the story that includes at least a portion of the tagged story model and the adaptable dialog set;
presenting, via a user interface presented on the first device, the information model for verification;
receiving, from the first device, user input indicating approval of the information model;
storing the information model; and
engaging, by a reading tool component associated with the reading comprehension tool, in a dialog session with a second user associated with a student account, wherein the dialog session uses the information model to evaluate the reading comprehension level of the second user.
2. The computer-implemented method as recited in claim 1 , further comprising engaging in the dialog session with the second user using the adaptable dialog set to evaluate a user understanding of the story based at least in part on the information model.
3. The computer-implemented method as recited in claim 2 , wherein engaging in the dialog session comprises:
selecting a first dialog from the adaptable dialog set;
presenting the first dialog;
receiving a response; and
analyzing the response to select a second dialog.
4. The computer-implemented method as recited in claim 1 , wherein the one or more labels include an enumerated value that increments based at least in part on element type.
5. The computer-implemented method as recited in claim 1 , wherein the adaptable dialog set includes one or more dialogs related to the one or more elements.
6. The computer-implemented method as recited in claim 1 , wherein generating the adaptable dialog set comprises:
retrieving a collection of dialogs; and
selecting a dialog from the collection of dialogs that has an element field that includes an element type associated with the one or more labels.
7. The computer-implemented method as recited in claim 6 , wherein determining the at least one answer is further based at least in part on replacing the element field with information tagged by the one or more labels.
8. The computer-implemented method as recited in claim 1 , wherein generating the adaptable dialog set comprises:
retrieving a collection of dialogs;
identifying a first dialog from the collection of dialogs that has an element field that includes an element type associated with the one or more labels, wherein the first dialog includes a first question template; and
determining a first answer to the first question template is based at least in part on replacing the element field with information tagged by the one or more labels.
9. The computer-implemented method as recited in claim 8 , further comprising:
identifying a second dialog including a second question template from the collection of dialogs; and
determining a second answer for the second question template based at least in part by analyzing the information tagged by the one or more labels.
10. The computer-implemented method as recited in claim 1 , further comprising training one or more machine learning models to process the story.
11. A system comprising:
one or more processors; and
programming instructions configured to be executed by the one or more processors to perform operations including:
receiving, from a first device associated with a first user, a story, wherein the first user is associated with a first account for a reading comprehension tool;
processing, by the reading comprehension tool, the story to identify one or more elements in the story, the one or more elements including at least one of a character, an event, or a setting associated with the story;
generating, by an authoring tool component associated with the reading comprehension tool, a tagged story model by tagging individual elements of the one or more elements with one or more labels;
generating, based at least in part on the tagged story model, an adaptable dialog set associated with the one or more elements, wherein the adaptable dialog set includes one or more questions for evaluating reading comprehension level;
determining, based at least in part on the one or more labels in the tagged story model, at least one answer for the one or more questions;
generating an information model of the story that includes at least a portion of the tagged story model and the adaptable dialog set; and
engaging, by a reading tool component associated with the reading comprehension tool, in a dialog session with a second user associated with a second account, wherein the dialog session uses the information model to evaluate the reading comprehension level of the second user.
12. The system as recited in claim 11 , the operations further comprising:
presenting, via a user interface presented on the first device, the information model for verification;
receiving, from the first device, user input indicating approval of the information model; and
storing the information model.
13. The system as recited in claim 11 , the operations further comprising:
presenting, via a user interface presented on the first device, the information model including the portion of the tagged story model;
receiving user input to tag an element of the one or more elements with a new label;
determining, based at least in part on the new label, an additional answer for the one or more questions; and
processing the tagged story model to generate a second learning model based at least in part on the new label.
14. The system as recited in claim 13 , the operations further comprising:
receiving a second story;
processing the second story to identify one or more second elements in the second story based at least in part on the second learning model; and
tagging at least one element of the one or more second elements with the new label to generate a tagged second story based at least in part on the second learning model.
15. The system as recited in claim 14 , the operations further comprising:
presenting, via the user interface, the tagged second story; and
receiving additional user input to add a dialog to the tagged second story.
16. The system as recited in claim 15 , the operations further comprising processing the tagged second story to generate a third learning model based at least in part on the dialog.
17. One or more non-transitory computer-readable media storing computer executable instructions that, when executed, cause one or more processors to perform operations comprising:
receiving, from a first device associated with a first user, a story, wherein the first user is associated with a first account for a reading comprehension tool;
processing, by the reading comprehension tool, the story to identify one or more elements in the story, the one or more elements including at least one of a character, an event, or a setting associated with the story;
generating, by an authoring tool component associated with the reading comprehension tool, a tagged story model by tagging individual elements of the one or more elements with one or more labels;
generating, by the authoring tool component and based at least in part on the tagged story model, an adaptable dialog set associated with the one or more elements, wherein the adaptable dialog set includes one or more questions for evaluating reading comprehension level;
determining, based at least in part on the one or more labels in the tagged story model, at least one answer for the one or more questions;
generating an information model of the story that includes at least a portion of the tagged story model and the adaptable dialog set; and
engaging, by a reading tool component associated with the reading comprehension tool, in a dialog session with a second user associated with a second account, wherein the dialog session uses the information model to evaluate the reading comprehension level of the second user.
18. The one or more non-transitory computer-readable media of claim 17 , the operations further comprising:
presenting, via a user interface presented on the first device, the information model for modification;
receiving, from the first device, user input for adding a new dialog to the adaptable dialog set; and
storing the information model.
19. The one or more non-transitory computer-readable media of claim 18 , wherein generating the adaptable dialog set comprises:
retrieving a collection of dialogs;
identifying a dialog from the collection of dialogs that has an element field including an element type associated with the one or more labels; and
replacing the element field with information tagged by the one or more labels.
20. The one or more non-transitory computer-readable media of claim 19 , wherein the user input includes adding a new label, and further comprising:
identifying a second dialog from the collection of dialogs that is associated with the element type associated with the new label; and
prompting for additional user input to add the second dialog.