USER-BASED EXTRACTION OF CONTENT
Disclosed are various embodiments for redacting or modifying content in documents that are provided to users. A user profile can be generated by analyzing user data within or external to the enterprise. A document can also be analyzed to identify or classify the document's components. A document or other content can be modified or redacted based upon the user profile and the analysis of the document.
1 . A non-transitory computer-readable medium embodying a program executable in a computing device, the program, when executed by the computing device, being configured to cause the computing device to at least:
identify a user in a user directory;
generate a user profile based upon an analysis of an email archive of the user, a group membership of the user within the user directory, a calendar archive of the user, or historical tracking data associated with the user;
identify a document accessible to the user;
tag a plurality of document components from the document based upon an analysis of a content of the document; and
generate a personalized presentation of the document based upon the plurality of the document components and the user profile.
2 . The non-transitory computer-readable medium of claim 1 , wherein the program generates the personalized presentation of the document by causing the computing device to at least:
identify a verbosity level based upon the historical tracking data, wherein the historical tracking data further comprises a reading profile of the user associated with a plurality of previous documents; and
redacting at least one document component from the document based upon the verbosity level.
3 . The non-transitory computer-readable medium of claim 1 , wherein the program generates the personalized presentation of the document by causing the computing device to at least:
generate at least one contextual hint corresponding to a document component from the document; and
cause the at least one contextual hint to be rendered in the personalized presentation alongside the document component.
4 . The non-transitory computer-readable medium of claim 1 , wherein the program generates the personalized presentation of the document by causing the computing device to at least:
generate a verbosity user interface component facilitating adjustment of the verbosity of the personalized presentation of the document; and
in response to a change in a selected verbosity level, causing the personalized presentation of the document to be updated.
5 . The non-transitory computer-readable medium of claim 1 , wherein the program generates the user profile based upon a further analysis of a plurality of documents associated with the user in a document archive.
6 . The non-transitory computer-readable medium of claim 1 , wherein the program tags the plurality of document components from the document based upon the analysis of a content of the document by utilizing a sequence-to-class deep learning model that takes a sequence of text as an input and returns a tag the represents the sequence of text.
7 . The non-transitory computer-readable medium of claim 1 , wherein the personalized presentation of the document based upon the plurality of the document components and the user profile is generated by a comparison module that is trained using the user profile as an input and the personalized presentation is generated using a reverse-feeding dictionary.
8 . A system, comprising:
at least one computing device;
at least one application executed by the at least one computing device, the at least one application, when executed, causing the at least one computing device to at least:
identify a user in a user directory;
generate a user profile based upon an analysis of an email archive of the user, a group membership of the user within the user directory, a calendar archive of the user, or historical tracking data associated with the user;
identify a document accessible to the user;
tag a plurality of document components from the document based upon an analysis of a content of the document; and
generate a personalized presentation of the document based upon the plurality of the document components and the user profile.
9 . The system of claim wherein the at least one application generates the personalized presentation of the document by causing the computing device to at least:
identify a verbosity level based upon the historical tracking data, wherein the historical tracking data further comprises a reading profile of the user associated with a plurality of previous documents; and
redacting at least one document component from the document based upon the verbosity level.
101 . The system of claim 8 , wherein the at least one application generates the personalized presentation of the document by causing the computing device to at least:
generate at least one contextual hint corresponding to a document component from the document; and
cause the at least one contextual hint to be rendered in the personalized presentation alongside the document component.
11 . The system of claim 8 , wherein the at least one application generates the personalized presentation of the document by causing the computing device to at least:
generate a verbosity user interface component facilitating adjustment of the verbosity of the personalized presentation of the document; and
in response to a change in a selected verbosity level, causing the personalized presentation of the document to be updated.
12 . The system of claim 8 , wherein the at least one application generates the user profile based upon a further analysis of a plurality of documents associated with the user in a document archive.
13 . The system of claim 8 , wherein the at least one application tags the plurality of document components from the document based upon the analysis of a content of the document by utilizing a sequence-to-class deep learning model that takes a sequence of text as an input and returns a tag the represents the sequence of text.
14 . The system of claim wherein the personalized presentation of the document based upon the plurality of the document components and the user profile is generated by a comparison module that is trained using the user profile as an input and the personalized presentation is generated using a reverse-feeding dictionary.
15 . A method, comprising:
obtaining, by at least one computing device, a request to provide a document to a user associated with a user account;
identify a user in a user directory;
generate a user profile based upon an analysis of an email archive of the user, a group membership of the user within the user directory, a calendar archive of the user, or historical tracking data associated with the user;
identify a document accessible to the user;
tag a plurality of document components from the document based upon an analysis of a content of the document; and
generate a personalized presentation of the document based upon the plurality of the document components and the user profile.
16 . The method of claim wherein generating the personalized presentation of the document further comprises:
identify a verbosity level based upon the historical tracking data, wherein the historical tracking data further comprises a reading profile of the user associated with a plurality of previous documents; and
redacting at least one document component from the document based upon the verbosity level.
17 . The method of claim 15 , wherein generating the personalized presentation of the document further comprises:
generate at least one contextual hint corresponding to a document component from the document; and
cause the at least one contextual hint to be rendered in the personalized presentation alongside the document component.
18 . The method of claim 15 , wherein generating the personalized presentation of the document further comprises:
generate a verbosity user interface component facilitating adjustment of the verbosity of the personalized presentation of the document; and
in response to a change in a selected verbosity level, causing the personalized presentation of the document to be updated.
19 . The method of claim 15 , wherein generating the user profile is based upon a further analysis of a plurality of documents associated with the user in a document archive.
20 . The method of claim 15 , wherein the personalized presentation of the document based upon the plurality of the document components and the user profile is generated by a comparison module that is trained using the user profile as an input and the personalized presentation is generated using a reverse-feeding dictionary.