System and method for analyzing partial utterances

A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to use a spoken language understanding system to:

• receive a user utterance;
• obtain a set of rules that define a plurality of check conditions corresponding to the detection of partial utterances;
• detect whether the user utterance is a partial utterance by comparing the user utterance with the set of rules to determine whether the user utterance matches a check condition;
• provide a detection advice code corresponding to the matched check condition if the user utterance is detected as a partial utterance, wherein the detection advice code includes at least one action to be taken in response to the user's utterance;
• classify the user utterance into a class of a plurality of classes if the user utterance is detected as a partial utterance, wherein the plurality of classes are different from the check conditions;
• provide a classification advice code corresponding to the class, wherein the classification advice code includes at least one action to be taken in response to the users utterance;
• execute a resolution strategy based on at least one of the detection advice code and the classification advice code;
• compare the detection advice code with the classification advice code;
• and
• wherein comparing the detection advice code with the classification advice code includes determining whether the detection advice code is the same as or different from the classification advice code;
• and
• wherein, when the detection code is determined to be different from the classification code, a class confidence score based on the probability that the user utterance fits within the class is analyzed, and the analysis of the class confidence score is used to determine which of the detection advice code and the classification advice code is sent to the dialogue management system.

A partial utterance analysis system, comprising:

• one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to:
• receive a user utterance;
• obtain a set of rules that define a plurality of check conditions corresponding to the detection of partial utterances;
• detect whether the user utterance is a partial utterance by comparing the user utterance with the set of rules to determine whether the user utterance matches a check condition;
• provide a detection advice code corresponding to the matched check condition if the user utterance is detected as a partial utterance, wherein the detection advice code includes at least one action to be taken in response to the user's utterance;
• classify the user utterance into a class of a plurality of classes if the user utterance is detected as a partial utterance, wherein the plurality of classes are different from the check conditions;
• provide a classification advice code corresponding to the class, wherein the classification advice code includes at least one action to be taken in response to the user's utterance;
• execute a resolution strategy based on at least one of the detection advice code and the classification advice code;
• compare the detection advice code with the classification advice code;
• and
• wherein comparing the detection advice code with the classification advice code includes determining whether the detection advice code is the same as or different from the classification advice code;
• and
• wherein, when the detection code is determined to be different from the classification code, a class confidence score based on the probability that the user utterance fits within the class is analyzed, and the analysis of the class confidence score is used to determine which of the detection advice code and the classification advice code is sent to the dialogue management system.