Query Interpretation via Device Orientation and Location
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users often need to know the name, pronunciation, and spelling of an entity to submit queries, which can be cumbersome, especially when interacting with search systems, and this requires multiple queries to retrieve information, increasing bandwidth and computational load.
Innovation Solution
The system interprets user queries by determining the approximate location and orientation of a user device, identifying a set of entities within that context, selecting one, and rewriting the query to explicitly reference the entity, allowing users to query without knowing the entity's name or proper spelling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users submit queries without knowing the entity name, then ease of operation is improved, but query interpretation accuracy deteriorates
Solution Approach 1:
The system performs preliminary actions by obtaining the user device's location and orientation data before processing the query. This allows the system to pre-identify candidate entities in the vicinity, so when an ambiguous query is received, the interpretation can be quickly resolved using the pre-computed entity context.
Solution Approach 2:
The patent introduces an intermediary entity resolution mechanism that acts as a mediator between the user's ambiguous query and the search system. By using location and orientation as intermediate parameters, the system translates vague queries into specific entity-referenced queries, resolving the accuracy issue while maintaining ease of use.
2Measurement precision
If the system processes ambiguous queries with location and orientation data, then query interpretation accuracy is improved, but computational load increases
Solution Approach 1:
Instead of processing all possible entities globally, the system applies local quality by focusing computational resources on entities within the user's specific location and orientation context. This localized approach processes only relevant candidate entities, significantly reducing the overall computational load while maintaining high interpretation accuracy.
Solution Approach 2:
The system performs partial action by obtaining location and orientation data only when needed for ambiguous queries, rather than continuously. This selective processing reduces unnecessary computational overhead while ensuring accurate interpretation when required.
3Measurement precision
If users must know entity names to query, then query accuracy is improved, but ease of operation deteriorates
Solution Approach 1:
The system enables self-service by automatically resolving entity references based on location and orientation data without requiring user input of entity names. The query interpretation system serves itself by inferring the intended entity from contextual data, eliminating the need for users to manually provide accurate entity identifiers.
Solution Approach 2:
By pre-obtaining location and orientation data, the system prepares the contextual information needed to interpret ambiguous queries accurately, allowing users to submit natural language queries without prior knowledge of entity names while maintaining high query accuracy.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving a query provided from a user device, and determining that the query implicitly references some entity, and in response: obtaining an approximate location and an approximate orientation of the user device, obtaining a set of entities including one or more entities, each entity in the set of entities being at least partially within a range that is based on the approximate location and that includes an extended portion based on the approximate orientation, selecting an entity from the set of entities, and providing a revised query based on the query and the entity, the revised query explicitly referencing the entity.