Back to results
US 12,737,644 B2Granted
Continuous learning of digital assistants
AssigneeSap SE
InventorsJulian Seibel, Steffen Terheiden
Life of this patent
Estimating the term…
Abstract
Various embodiments for a continuous learning system are described herein. An embodiment operates by receiving a query from a user and identifying an unknown phrase in the query. User feedback regarding the unknown phrase is requested and received. A first pre-existing entity of a plurality of pre-existing entities that corresponds to the received user feedback is identified. A relationship between the first pre-existing entity and the unknown is added to the knowledgebase. The query is executed against the knowledgebase using the first pre-existing entity. A response, to the executed query, is provided to the user.
Claims20
A method comprising:
- receiving a query from a user;
- identifying an unknown phrase in the query, wherein the unknown phrase does not exist in a knowledgebase comprising a plurality of pre-existing entities;
- requesting, by one or more processors, feedback from the user regarding the unknown phrase;
- receiving, by the one or more processors, the feedback regarding the unknown phrase;
- parsing, by a natural language model, the feedback regarding the unknown phrase to identify non-extraneous words that define the unknown phrase;
- identifying, by the one or more processors, a relationship between a first pre-existing entity of the plurality of pre-existing entities and the unknown phrase based on the non-extraneous words, wherein the relationship defines the unknown phrase;
- requesting, by the one or more processors, confirmation that the relationship between the first pre-existing entity and the unknown phrase is accurate based on additional user feedback from a set of users, wherein the set of users are in a same group-specific knowledgebase extension as the user;
- receiving, by the one or more processors, the requested confirmation from the set of users that the relationship between the first pre-existing entity and the unknown phrase is accurate;
- adding, by the one or more processors, the relationship between the first pre-existing entity and the unknown phrase to the knowledgebase based on the confirmation;
- executing, by the one or more processors, the query against the knowledgebase using the first pre-existing entity;
- and
- providing, by the one or more processors, a response, to the executed query, to the user.
9 more elements
A system comprising at least one processor, the at least one processor configured to perform operations comprising:
- receiving a query from a user;
- identifying an unknown phrase in the query, wherein the unknown phrase does not exist in a knowledgebase comprising a plurality of pre-existing entities;
- requesting feedback from the user regarding the unknown phrase;
- receiving the feedback regarding the unknown phrase;
- parsing, by a natural language model, the feedback regarding the unknown phrase to identify non-extraneous words that define the unknown phrase;
- identifying a relationship between a first pre-existing entity of the plurality of pre-existing entities and the unknown phrase based on the non-extraneous words, wherein the relationship defines the unknown phrase;
- requesting confirmation that the relationship between the first pre-existing entity and the unknown phrase is accurate based on additional user feedback from a set of users, wherein the set of users are in a same group-specific knowledgebase extension as the user;
- receiving the requested confirmation from the set of users that the relationship between the first pre-existing entity and the unknown phrase is accurate;
- adding the relationship between the first pre-existing entity and the unknown phrase to the knowledgebase based on the confirmation;
- executing the query against the knowledgebase using the first pre-existing entity;
- and
- providing a response, to the executed query, to the user.
9 more elements
A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:
- receiving a query from a user;
- identifying an unknown phrase in the query, wherein the unknown phrase does not exist in a knowledgebase comprising a plurality of pre-existing entities;
- requesting feedback from the user regarding the unknown phrase;
- receiving the feedback regarding the unknown phrase;
- parsing, by a natural language model, the feedback regarding the unknown phrase to identify non-extraneous words that define the unknown phrase;
- identifying a relationship between a first pre-existing entity of the plurality of pre-existing entities and the unknown phrase based on the non-extraneous words, wherein the relationship defines the unknown phrase;
- requesting confirmation that the relationship between the first pre-existing entity and the unknown phrase is accurate based on additional user feedback from a set of users, wherein the set of users are in a same group-specific knowledgebase extension as the user;
- receiving the requested confirmation from the set of users that the relationship between the first pre-existing entity and the unknown phrase is accurate;
- adding the relationship between the first pre-existing entity and the unknown phrase to the knowledgebase based on the confirmation;
- executing the query against the knowledgebase using the first pre-existing entity;
- and
- providing a response, to the executed query, to the user.
9 more elements
Dependencies are inferred from the claim text.