Utilizing machine learning models to identify insights in a document
A device receives document information associated with a document, and receives a request to identify insights in the document information. The device performs, based on the request, natural language processing on the document information to identify words, phrases, and sentences in the document information, and utilizes a first machine learning model with the words, the phrases, and the sentences to identify information indicating abstract insights, concrete insights, and non-insights in the document. The device utilizes a second machine learning model to match the abstract insights with particular concrete insights that are different than the concrete insights, and utilizes a third machine learning model to determine particular insights based on the non-insights. The device generates an insight document that includes the concrete insights, the abstract insights matched with the particular concrete insights, and the particular insights determined based on the non-insights.
1. A method comprising:
performing, by a device and based on a user interface element being selected to provide document insights, natural language processing on a document to identify one or more alphabetic words, alphanumeric words, numeric words, phrases, or sentences in the document;
utilizing, by the device, a first machine learning model with the one or more alphabetic words, alphanumeric words, numeric words, phrases, or sentences to identify information indicating an abstract insight, a concrete insight, and a non-insight in the document;
utilizing, by the device, a second machine learning model to match the abstract insight with a particular concrete insight that is different than the concrete insight;
utilizing, by the device, a third machine learning model to match a particular insight in a repository with the non-insight;
generating, by the device, an insight document that includes information indicating the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight;
determining, by the device and based on the insight document, a recommended document; and
providing, by the device and for display, information identifying the recommended document.
2. The method of claim 1 , where performing the natural language processing on the document comprises:
performing the natural language processing on the document based on at least one of:
a user history,
a type of the document, or
a time of day.
3. The method of claim 1 , where the particular concrete insight is stored in at least one of the repository or another repository.
4. The method of claim 1 , further comprising:
assigning scores to the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight;
ranking the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight based on the scores; and
including a ranked list of insights in the insight document based on ranking the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight based on the scores.
5. The method of claim 1 , further comprising:
comparing information provided in the insight document and one or more documents stored in a document repository; and
matching the information provided in the insight document with the recommended document; and
where providing, for display, the information identifying the recommended document comprises:
providing, for display, the information identifying the recommended document based on matching the information provided in the insight document with the recommended document.
6. The method of claim 1 , further comprising:
providing, for display, the recommended document.
7. The method of claim 1 , further comprising:
providing the abstract insight, the concrete insight, and the non-insight to at least one of the first machine learning model, the second machine learning model, or the third machine learning model to train at least one of the first machine learning model, the second machine learning model, or the third machine learning model.
8. A device, comprising:
one or more memories; and
one or more processors communicatively coupled to the one or more memories, configured to:
perform, based on a user interface element being selected to provide document insights, natural language processing on a document to identify one or more alphabetic words, alphanumeric words, numeric words, phrases, or sentences in the document;
utilize a first machine learning model with the one or more alphabetic words, alphanumeric words, numeric words, phrases, or sentences to identify information indicating an abstract insight, a concrete insight, and a non-insight in the document;
utilize a second machine learning model to match the abstract insight with a particular concrete insight that is different than the concrete insight;
utilize a third machine learning model to match a particular insight in a repository with the non-insight;
generate an insight document that includes information indicating the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight;
determine, based on the insight document, a recommended document; and
provide, for display, information identifying the recommended document.
9. The device of claim 8 , where the one or more processors, when performing the natural language processing on the document, are configured to:
perform the natural language processing on the document based on user settings.
10. The device of claim 8 , where the document is a first document; and
where the one or more processors are further configured to:
automatically perform the natural language processing on a second document based on historical information.
11. The device of claim 8 , where the one or more processors, when performing the natural language processing on the document, are configured to:
perform the natural language processing on the document based on a type of the document.
12. The device of claim 8 , where the one or more processors are further configured to:
compare information provided in the insight document and one or more documents stored in a document repository; and
match the information provided in the insight document with the recommended document; and
where the one or more processors, when providing, for display, the information identifying the recommended document, are configured to:
provide, for display, the information identifying the recommended document based on matching the information provided in the insight document with the recommended document.
13. The device of claim 8 , where the first machine learning model includes one or more of:
a decision tree learning model,
a learner classifier systems model,
a nearest neighbor model,
a support vector machine model,
a clustering model,
a neural network model, or
a latent variable model.
14. The device of claim 8 , where the one or more processors are further configured to:
provide the abstract insight, the concrete insight, and the non-insight to at least one of the first machine learning model, the second machine learning model, or the third machine learning model to train at least one of the first machine learning model, the second machine learning model, or the third machine learning model.
15. A non-transitory computer-readable medium storing instructions, the instructions comprising:
one or more instructions that, when executed by one or more processors, cause the one or more processors to:
perform, based on a user interface element being selected to provide document insights, natural language processing on a document to identify one or more alphabetic words, alphanumeric words, numeric words, phrases, or sentences in the document;
utilize a first machine learning model with the one or more alphabetic words, alphanumeric words, numeric words, phrases, or sentences to identify information indicating an abstract insight, a concrete insight, and a non-insight in the document;
utilize a second machine learning model to match the abstract insight with a particular concrete insight that is different than the concrete insight;
utilize a third machine learning model to match a particular insight in a repository with the non-insight;
generate an insight document that includes information indicating the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight;
determine, based on the insight document, a recommended document; and
provide, for display, information identifying the recommended document.
16. The non-transitory computer-readable medium of claim 15 , where the one or more instructions, that cause the one or more processors to perform the natural language processing on the document, cause the one or more processors to:
perform the natural language processing on the document based on at least one of:
a user history,
a type of the document, or
a time of day.
17. The non-transitory computer-readable medium of claim 15 , where the particular concrete insight is stored in at least one of the repository or another repository.
18. The non-transitory computer-readable medium of claim 15 , where the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
assign scores to the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight;
rank the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight based on the scores; and
include a ranked list of insights in the insight document based on ranking the concrete insight, the abstract insight matched with the particular concrete insight, and the particular insight based on the scores.
19. The non-transitory computer-readable medium of claim 15 , where the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
compare information provided in the insight document and one or more documents stored in a document repository; and
match the information provided in the insight document with the recommended document; and
where the one or more instructions, that cause the one or more processors to provide, for display, the information identifying the recommended document, cause the one or more processors to:
provide, for display, the information identifying the recommended document based on matching the information provided in the insight document with the recommended document.
20. The non-transitory computer-readable medium of claim 15 , where the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
provide the abstract insight, the concrete insight, and the non-insight to at least one of the first machine learning model, the second machine learning model, or the third machine learning model to train at least one of the first machine learning model, the second machine learning model, or the third machine learning model.