Hierarchical Ontology for E-commerce Voice Query Parsing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voice assistant devices struggle to accurately interpret complex queries, including those with multiple intents, negations, and ambiguous entities, in e-commerce transactions.
Innovation Solution
The implementation of a hierarchical ontology system that generates a semantic representation of user queries, comprising ecommerce actions, entities, and properties, allowing for the parsing and understanding of complex queries by associating user inputs with actions, entities, and properties, and determining syntactic and semantic dependencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current voice assistant devices use simple intent and entity identification methods, then the device complexity is low, but the measurement precision of query interpretation is insufficient for complex queries
Solution Approach 1:
The patent segments the query interpretation process into distinct components: intent identification, entity recognition, and relationship extraction. Each component handles a specific aspect of the query, allowing the system to process complex multi-intent queries by breaking them down into manageable parts that can be analyzed separately and then integrated.
Solution Approach 2:
The patent introduces an intermediary semantic representation layer that bridges the gap between raw query text and actionable commands. This intermediate representation captures the meaning and relationships in the query before executing actions, enabling more accurate interpretation of complex queries while maintaining a structured processing pipeline.
2Reliability
If the voice assistant device uses basic parsing methods, then the ease of operation is maintained, but the reliability of handling complex queries with multiple intents and negations deteriorates
Solution Approach 1:
The patent performs preliminary actions by pre-defining intent categories, entity types, and relationship patterns before query processing. This preparation work enables the system to reliably handle complex queries by matching them against known patterns, improving reliability without requiring complex real-time processing logic.
Solution Approach 2:
The patent implements feedback mechanisms where the semantic representation is validated and refined through multiple processing stages. The system checks for consistency in multi-intent queries, resolves ambiguities through iterative analysis, and adjusts the interpretation based on contextual clues, thereby improving reliability while maintaining operational simplicity.
3Productivity
If the system uses detailed semantic representation with hierarchical ontology, then the productivity of query processing is improved, but the loss of information in representing ambiguous entities increases
Solution Approach 1:
The patent resolves entity ambiguity by adding temporal and contextual dimensions to the representation. Instead of losing ambiguous entity information, the system captures multiple possible interpretations and their relationships to other query elements, allowing disambiguation based on context from the full query and user history.
Solution Approach 2:
The patent uses nested hierarchical ontology structures where entities can contain multiple levels of specificity. Ambiguous entities are represented as nested structures that preserve both the general category and specific instances, allowing the system to maintain detailed information while organizing it in a computationally efficient hierarchical format.
Data Source
AI summary
An e-commerce voice representation system is provided and generally includes a server, a database, and a customer computing device. The server may generate a hierarchical ontology to semantically represent user queries generated by the customer computing device. The server may receive from the customer computing device data representing an e-commerce user query. The server may parse the e-commerce user query based on syntactic dependencies between the words in the e-commerce user query and translating those dependencies into semantic relations between the words. The received e-commerce user query may be tagged with actions, entities, and properties of the hierarchical ontology. Identified products may be linked to conceptual classes of the hierarchical ontology. Syntactic dependencies between words of the e-commerce user query may then be determined, and a semantic representation of the e-commerce user query may be generated based on the syntactic dependencies and mapping rules.