Natural Language Query Parameter Mapping for Ambiguous Search Intent
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Natural language searches are rudimentary, requiring word or phrase matching and are unintuitive, making it difficult to transform complex human utterances into query parameters for database queries.
Innovation Solution
A system and method that tokenizes utterances into feature vectors, assigns entity labels, and resolves them to standardized database query values using machine learning models and string-based algorithms, including disambiguation engines to handle colloquialisms and synonyms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If natural language searches use simple word or phrase matching, then the search operation is fast and simple, but the search capability is rudimentary and unintuitive
Solution Approach 1:
The patent introduces an intermediary processing layer between natural language input and database query execution. This layer includes tokenization, feature vector transformation, entity labeling, and disambiguation components that translate colloquial language into standardized query parameters, enabling both natural language convenience and precise database searching
Solution Approach 2:
The patent replaces traditional mechanical keyword-matching search mechanisms with machine learning-based semantic understanding systems. Neural network models analyze the meaning and context of natural language utterances, transforming them into structured query parameters without requiring rigid keyword correspondence
2Measurement precision
If complex processing is applied to transform utterances into query parameters, then the search accuracy improves, but the processing time and system complexity increase
Solution Approach 1:
The patent implements preliminary action by pre-training machine learning models on large corpora of natural language and query parameter mappings. During inference, these pre-trained models rapidly transform new utterances into query parameters without requiring complex real-time processing, as the transformation rules have been learned in advance
Solution Approach 2:
The patent segments the complex transformation process into independent modular components: tokenization, feature extraction, entity recognition, disambiguation, and query parameter generation. Each component processes a specific aspect of the utterance, allowing parallel processing and reducing overall computation time while maintaining high accuracy
3Manufacturing precision
If UI widgets are used for specifying search categories and date windows, then the query parameters can be precisely controlled, but the user interaction becomes time consuming and unintuitive
Solution Approach 1:
The patent enables self-service by allowing the system to automatically extract and interpret search parameters directly from natural language utterances without requiring users to manually interact with UI widgets. The machine learning system autonomously identifies entities, determines their semantic roles, and converts them into properly formatted query parameters, saving user time while maintaining precision
Data Source
AI summary
A method may receive a text string of an utterance. A method may tokenize the utterance into a plurality of tokens. A method may transform the plurality of tokens into a plurality of feature vectors. A method may assign an entity label to each of the plurality of feature vectors. A method may resolve each feature vector of the plurality of feature vectors to a corresponding standardized value of a database query language.