Intelligent interpretation of temporal expressions
AssigneeMicrosoft Technology Licensing, LLC
InventorsPamela Bhattacharya, Christopher Alan Meek, Oleksandr Polozov, Alex James Boyd
Life of this patent
Estimating the term…
Abstract
A method and system for predicting an intended time interval for a content segment may include receiving a request for natural language processing (NLP) of the content segment, the content segment including one or more temporal expressions, accessing contextual data associated with each of the one or more temporal expressions, decoding the content segment into a program that describes a temporal logic of the content segment based on the one or more temporal expressions, evaluating the program using the contextual data to predict an intended time interval for the content segment, and providing the intended time interval as an output.
Claims15
A data processing system comprising:
- a processor;
- and
- a memory in communication with the processor, the memory comprising executable instructions that, when executed by, the processor, cause the data processing system to perform functions of:
- receiving a request for natural language processing (NLP) of a content segment, the content segment including one or more temporal expressions;
- accessing contextual data associated with each of the one or more temporal expressions;
- parsing the content segment to identify one or more tokens in the content segment
- generating predicted program tokens for an instance of a program in a domain-specific language;
- generating the program in the domain-specific language, using an NLP model and based on the one or more tokens of the content segment and the predicted program tokens, the program describing a temporal logic of the content segment based on the one or more temporal expressions;
- providing the program and contextual data as inputs to an interpreter;
- executing the program using the interpreter to predict an intended time interval for the content segment;
- and
- providing the intended time interval as an output,
- wherein the NLP model is trained by using at least one of supervised learning and reinforcement learning.
A method for predicting an intended time interval for a content segment, comprising:
- receiving a request for natural language processing (NLP) of the content segment, the content segment including one or more temporal expressions;
- accessing contextual data associated with each of the one or more temporal expressions;
- parsing the content segment to identify one or more tokens in the content segment;
- generating predicted program tokens for an instance of a program in a domain-specific language;
- generating the program in the domain-specific language, using an NLP model and based on the one or more tokens of the content segment and the predicted program tokens, the program describing a temporal logic of the content segment based on the one or more temporal expressions;
- providing the program and contextual data as inputs to an interpreter;
- executing the program using the interpreter to predict an intended time interval for the content segment;
- and
- providing the intended time interval as an output,
- wherein the NLP model is trained by using at least one of supervised learning and reinforcement learning.
A non-transitory computer readable medium on which are stored instructions that, when executed, cause a programmable device to:
- receive a request for natural language processing (NLP) of a content segment, the content segment including one or more temporal expressions;
- access contextual data associated with each of the one or more temporal expressions;
- parse the content segment to identify one or more tokens in the content segment;
- generate predicted program tokens for an instance of a program in a domain-specific language;
- generate the program in the domain-specific language, using an NLP model and based on the one or more tokens of the content segment and the predicted program tokens, the program describing a temporal logic of the content segment based on the one or more temporal expressions;
- provide the program and contextual data as inputs to an interpreter;
- execute the program using the interpreter to predict an intended time interval for the content segment;
- and
- provide the intended time interval as an output,
- wherein the NLP model is trained by using at least one of supervised learning and reinforcement learning.
Dependencies are inferred from the claim text.