17818708. SYSTEMS AND METHODS FOR HIERARCHICAL MULTI-LABEL MULTI-CLASS INTENT CLASSIFICATION simplified abstract (Capital One Services, LLC)
SYSTEMS AND METHODS FOR HIERARCHICAL MULTI-LABEL MULTI-CLASS INTENT CLASSIFICATION
Organization Name
Inventor(s)
Isha Chaturvedi of Mountain View CA (US)
SYSTEMS AND METHODS FOR HIERARCHICAL MULTI-LABEL MULTI-CLASS INTENT CLASSIFICATION - A simplified explanation of the abstract
This abstract first appeared for US patent application 17818708 titled 'SYSTEMS AND METHODS FOR HIERARCHICAL MULTI-LABEL MULTI-CLASS INTENT CLASSIFICATION
Simplified Explanation
Methods and systems for efficiently labeling user utterances and identifying novel user intents for large amounts of data using a machine learning model trained on embeddings of utterance data.
- Machine learning model trained on embeddings of utterance data
- Utilizes prototypical networks and hierarchical local binary classification
- Hierarchical multi-label multi-class classification approach
Potential Applications
- Customer service chatbots
- Voice assistants
- Social media sentiment analysis
Problems Solved
- Efficiently labeling user utterances
- Identifying novel user intents
- Handling large amounts of data in conversational interactions
Benefits
- Improved accuracy in labeling user utterances
- Enhanced understanding of user intents
- Scalability for processing large datasets
Original Abstract Submitted
Methods and systems are described herein for efficiently labeling user utterances, which may encompass any communication received from a user within a conversational interaction, and identifying novel user intents for large amounts of data. A machine learning model may be used, which is trained on embeddings of utterance data, and which may employ methods like prototypical networks and hierarchical local binary classification for hierarchical multi-label multi-class classification.
