Method and system for updating an intent space and estimating intent based on an intent space
AssigneeOath Inc.
InventorsZornitsa Kozareva, Scott Gaffney
Life of this patent
Estimating the term…
Abstract
The present teaching relates to updating an intent space and estimating intent based on an intent space. In one example, an initial intent space is obtained. Each intent in the initial intent space is characterized in one or more dimensions. At least one model is received. Each of the at least one model provides features in each of the dimensions and relationship thereof. A new intent associated with an intent in the initial intent space is determined based on the at least one model. Based on the new intent, the initial intent space is updated to derive an updated intent space.
Claims11
A method, implemented on a computing device having at least one processor, storage, and a communication platform capable of connecting to a network for estimating intent based on an intent space having a plurality of dimensions, the method comprising:
- receiving an input from a person;
- receiving at least one model, each of which characterizes each of the plurality of dimensions based on a set of features and obtains a closeness in relationship between each pair of features in the dimension by computing a distance between the pair of features mapped in a feature space;
- analyzing the input to extract one or more features in at least one of the plurality of dimensions in the intent space;
- obtaining contextual information related to the input;
- and
- determining, based on at least one of the distance, the contextual information, and the extracted one or more features, an intent from the intent space which includes a plurality of intents, each intent being added to the intent space based on a criterion corresponding to a frequency of occurrence of the intent being satisfied, wherein the determining further comprises:
- determining whether a number of the extracted one or more features from the input equals to a number of the plurality of dimensions,
- identifying, in response to a result of the determining being positive, the intent in the intent space based on the extracted one or more features from the input,
- identifying, in response to the result of the determining being negative, a plurality of intent candidates from the intent space based on the at least one model and the extracted one or more features, and
- selecting the intent from the plurality of intent candidates based on the contextual information.
A non-transitory, machine-readable medium having information recorded thereon for estimating intent based on an intent space having a plurality of dimensions, wherein the information, when read by a machine, causes the machine to perform the steps of:
- receiving an input from a person;
- receiving at least one model, each of which characterizes each of the plurality of dimensions based on the set of features and obtains a closeness in relationship between each pair of features in the dimension by computing a distance between the pair of features mapped in a feature space;
- analyzing the input to extract one or more features in at least one of the plurality of dimensions in the intent space;
- obtaining contextual information related to the input;
- determining, based on at least one of the distance, the contextual information, and the extracted one or more features, an intent from the intent space which includes a plurality of intents, each intent being added to the intent space based on a criterion corresponding to a frequency of occurrence of the intent being satisfied, wherein the determining further comprises:
- determining whether a number of the extracted one or more features from the input equals to a number of the plurality of dimensions,
- identifying, in response to a result of the determining being positive, the intent in the intent space based on the extracted one or more features from the input,
- identifying, in response to the result of the determining being negative, a plurality of intent candidates from the intent space based on the at least one model and the extracted one or more features, and
- selecting the intent from the plurality of intent candidates based on the contextual information.
Dependencies are inferred from the claim text.