Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12417355-B2
Publication Date
2025-09-16
Expiration Date
Abstract
Disambiguating user utterances is provided. Disambiguation of an utterance of a user is performed using up to a defined number of user intents in a set of possible user intents having highest confidence scores between a first confidence score threshold level and a second confidence score threshold level in response to determining that each user intent in the set of possible user intents does have a corresponding confidence score less than the second confidence score threshold level. Each of the up to the defined number of user intents is located in a user intent mapping table. A human interpretable label corresponding to each of the up to the defined number of user intents located in the user intent mapping table is extracted. Extracted human interpretable labels are assembled into a set of user intent options. The set of user intent options is sent to a client device of the user.
Core Innovation
The invention disambiguates user utterances using a chatbot by comparing a first plurality of user intents in a user intent file with a second plurality of user intents in a user intent mapping table. Based on the comparison, it generates a list of user intents present in the user intent file that are not present in the user intent mapping table and stores the list in the user intent mapping table.
The disambiguation is performed using a defined number of user intents selected from a set of possible user intents having highest confidence scores between a first confidence score threshold level and a second confidence score threshold level. The disambiguation is performed in response to a determination that each user intent in the set of possible user intents has a corresponding confidence score less than the second confidence score threshold level.
The invention determines a first user intent of the defined number of user intents is missing from the user intent mapping table when a confidence score corresponding to the first user intent is greater than or equal to the first confidence score threshold level and less than or equal to the second confidence score threshold level. Responsive to the determination, it generates a human interpretable label for the first user intent by modifying a user intent of the user intent file, updates the user intent mapping table, and sends the set of user intent options to a client device of the user.
Claims Coverage
The document provides independent claims directed to disambiguating user utterances using a computer-implemented method, a computer system, and a computer program product, plus a separate independent method variant. Across these claims, the inventive features center on confidence-score-threshold driven intent disambiguation, maintaining and updating a user intent mapping table, and generating human interpretable intent labels for presenting selectable intent options to a client device.
Comparing user intents in a user intent file with a user intent mapping table and storing missing intents
performing a comparison of a first plurality of user intents in a user intent file of a chatbot with a second plurality of user intents in a user intent mapping table, generating based on the comparison a list of user intents present in the first plurality that are not present in the second plurality, and storing the list in the user intent mapping table
Confidence-threshold-range driven disambiguation using highest-confidence intents
performing disambiguation of a user utterance based on a defined number of user intents in a set of possible user intents having highest confidence scores between a first confidence score threshold level and a second confidence score threshold level, wherein the disambiguation is performed responsive to a determination that each user intent in the set of possible user intents has a corresponding confidence score less than the second confidence score threshold level
Generating human interpretable labels for missing in-threshold user intents by modifying the chatbot user intent
determining a first user intent of the defined number is missing from the user intent mapping table when a confidence score corresponding to the first user intent is greater than or equal to the first confidence score threshold level and less than or equal to the second confidence score threshold level; generating responsive to the determination a human interpretable label for the first user intent by modifying a user intent of the user intent file of the chatbot
Assembling and sending user intent options with human interpretable labels to a client device
locating each user intent of the defined number in the user intent mapping table; extracting the human interpretable label corresponding to each located user intent; assembling a set of human interpretable labels into a set of user intent options; and sending the set of user intent options to a client device of the user
Across the independent claims, the core claimed coverage is a chatbot disambiguation approach that compares a user intent file to a user intent mapping table, disambiguates a user utterance using highest-confidence intents between first and second confidence score threshold levels, detects missing in-threshold intents, generates human interpretable labels by modifying the chatbot user intents, updates the mapping table, assembles labeled user intent options, and sends the options to a client device.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?