Textual summaries in information systems based on personalized prior knowledge
Examples of the present disclosure describe systems and methods for providing textual summaries based on personalized prior knowledge. In examples, a user request for a summary of document or an entity is received. The document or documents associated with the entity are separated into segments and semantic embeddings are created for each segment. The semantic embeddings, a context of the user request, and relevant information available in the requesting user's previous knowledge base are provided as input to a personal knowledge system. Based on the input, the personal knowledge system outputs an indication of the segments that should be summarized. The indicated segments are provided to a summarization system. The summarization system generates a document summary or an entity summary and provides the summary to the user.
1 . A system comprising:
a processing unit; and
memory coupled to the processing unit, the memory comprising computer executable instructions that, when executed, perform operations comprising:
receiving, from a user, a user request for a summary of a document;
identifying a context for the user request, the context comprising at least one of:
reading speed of the user; or
a determined amount of current availability for the user;
separating the document into segments;
generating semantic embeddings for the segments;
providing the semantic embeddings and the context to a knowledge system that is personalized for the user;
comparing each of the semantic embeddings to knowledge information for the user to determine a knowledge level of the user with topics represented by content in each of the semantic embeddings, wherein the knowledge level is based on a degree of similarity between the semantic embeddings and the knowledge information, wherein the knowledge information comprises:
objects interacted with by the user;
relationships between the objects; and
metadata associated with the objects and the relationships;
generating, by the knowledge system, a summarization determination for the document based on the context and the knowledge level of the user with the topics, wherein the summarization determination identifies which of the content in each of the semantic embeddings is to be summarized and an amount by which the content in each of the semantic embeddings is to be summarized, and wherein the summarization determination comprises computer instructions for converting the document into the summary of the document;
applying a presentation option to the summarization determination based on the context, wherein the presentation option prioritizes presentation of the topics based on the knowledge level of the user with the topics;
generating the summary of the document based on the summarization determination and the presentation option; and
providing the summary of the document to the user.
2 . The system of claim 1 , wherein the user request is an implicit user request that is determined based on user signals received from a user device of the user.
3 . The system of claim 2 , wherein the implicit user request corresponds to:
receiving the document via an electronic communication;
receiving a semantic reference to the document; or
a hover event on a link or an icon of the document.
4 . The system of claim 1 , wherein the context for the user request is identified using information associated with the user and at least one of:
information associated with a user device of the user; or
information associated with the document.
5 . The system of claim 4 , wherein the information associated with the user comprises at least one of:
behavioral tendencies of the user;
configuration settings or user preferences associated with the user; or
knowledge objectives indicated by the user.
6 . The system of claim 4 , wherein the information associated with the user device of the user comprises at least one of:
geographical location data for the user;
destination data for the user; or
capabilities of the user device.
7 . The system of claim 4 , wherein the information associated with the document comprises at least one of:
an average reading time for the document;
a reading complexity of the document; or
topics discussed in the document.
8 . The system of claim 4 , wherein:
the system is an enterprise system, the user being a member of the enterprise system; and
the context is identified using data sources of the enterprise system.
9 . The system of claim 1 , wherein the segments correspond to:
sections of the document;
paragraphs of the document; or
sentences of the document.
10 . The system of claim 1 , wherein the semantic embeddings represent feature vectors that comprises semantic information for a corresponding segment.
11 . The system of claim 1 , wherein the knowledge information for the user is provided to the knowledge system as part of providing the semantic embeddings and the context to the knowledge system.
12 . The system of claim 11 , wherein the knowledge information further comprises at least one of:
previously collected user signals for the user; or
existing user knowledge of the user.
13 . The system of claim 11 , wherein the knowledge system compares the semantic embeddings to the knowledge information to determine similarities between the semantic embeddings and the knowledge information.
14 . The system of claim 13 , wherein the similarities between the semantic embeddings and the knowledge information indicate whether the user is knowledgeable regarding topics or facts represented by the semantic embeddings.
15 . A method comprising:
receiving, from a user, a user request for a summary of a document;
identifying a context for the user request, the context comprising at least one of:
reading speed of the user; or
a determined amount of current availability for the user;
separating the document into segments;
generating semantic embeddings for the segments;
providing the semantic embeddings and the context to a knowledge system that is personalized for the user;
comparing each of the semantic embeddings to knowledge information for the user to determine a knowledge level of the user with topics represented by content in each of the semantic embeddings, wherein the knowledge information comprises:
objects interacted with by the user;
relationships between the objects; and
metadata associated with the objects and the relationships;
generating, by the knowledge system, a summarization determination for the document based on the context and the knowledge level of the user with the topics, wherein the summarization determination identifies which of the content in each of the semantic embeddings is to be summarized and an amount by which the content in each of the semantic embeddings is to be summarized, and wherein the summarization determination comprises computer instructions for converting the document into the summary of the document;
generating the summary of the document based on the summarization determination; and
providing the summary of the document to the user.
16 . The method of claim 15 , wherein generating the summarization determination comprises:
determining a summarization scope for the summary of the document, the summarization scope indicating that summary information for at least a portion of the semantic embeddings is to be omitted from the summary of the document.
17 . The method of claim 16 , wherein generating the summarization determination further comprises:
determining presentation options for the summary of the document, the presentation options indicating at least one of:
a presentation order for the segments; or
an output mode for the summary of the document.
18 . The method of claim 15 , wherein generating the summary of the document comprises:
identifying the segments corresponding to the semantic embeddings;
applying the summarization determination to the segments to generate modified segments; and
using the modified segments to generate the summary of the document.
19 . The method of claim 15 , wherein generating the summary of the document comprises:
accessing the document or a copy of the document; and
applying the summarization determination to the document or the copy of the document, wherein the summarization determination includes instructions for converting the document or the copy of the document into the summary of the document.
20 . A computing environment comprising:
a processing unit; and
memory coupled to the processing unit, the memory comprising computer executable instructions that, when executed, perform operations comprising:
receiving, from a user, a user request for a summary of a document, wherein the document comprises content on one or more topic areas;
identifying a context for the user request, wherein the context is based on at least one of:
reading speed of the user; or
a determined amount of current availability for the user;
generating semantic embeddings for the document;
providing, to a knowledge system, the semantic embeddings, the context, and knowledge information for the user, the knowledge information identifying a level of knowledge of the user for one or more areas of knowledge;
comparing each of the semantic embeddings to knowledge information for the user to determine a knowledge level of the user with topics represented by content in each of the semantic embeddings, wherein the knowledge information comprises:
objects interacted with by the user;
relationships between the objects; and
metadata associated with the objects and relationships;
generating, by the knowledge system, a summarization determination for the document based on the context and the knowledge level of the user with the topics, wherein the summarization determination identifies which of the content in each of the semantic embeddings is to be summarized and an amount by which the content in each of the semantic embeddings is to be summarized, and wherein the summarization determination comprises computer instructions for converting the document into the summary of the document;
applying a presentation option to the summarization determination based on the context;
generating the summary of the document based on the summarization determination and the presentation option; and
providing the summary of the document to the user.