Geographical Knowledge Graph for Map Search Query Interpretation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current map-based search systems fail to accurately provide query completion suggestions and search results that align with the user's intent, as they lack efficient methods to interpret and rank search queries based on geographical context and user intent.
Innovation Solution
The system employs a geographical knowledge graph that interprets map-based search queries by tagging terms as concepts, categories, attributes, or geographical entities, using a weighted finite state transducer and machine learning models to determine weight values and scores for each interpretation, thereby providing relevant completion suggestions and search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional map-based search systems are used, then the system structure is simple, but the accuracy of query completion suggestions and search results does not align with user intent
Solution Approach 1:
The patent segments the search query processing into distinct components: a geographical knowledge graph for structured geographical information, a weighted finite state transducer for multi-level tagging (geographical entities, concepts, categories, attributes), and machine learning models for scoring. This segmentation allows each component to specialize in specific tasks, improving overall accuracy while managing complexity through modular architecture.
Solution Approach 2:
The patent introduces a geographical knowledge graph as an intermediary layer between the user's search query and the search results. This knowledge graph acts as a mediator that structures geographical information and enables the weighted finite state transducer to interpret queries with geographical context, thereby improving accuracy without directly increasing the complexity of the core search engine.
2Reliability
If geographical context and user behavior data are integrated, then the relevance of search results improves, but the data processing complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-processing and structuring geographical information into a knowledge graph before search queries are submitted. User behavior data is also pre-analyzed to inform the machine learning models. This preliminary structuring reduces the complexity of real-time data processing during search operations, as the system only needs to query and score against pre-organized data structures.
Solution Approach 2:
The patent transforms unstructured geographical data and user behavior data into structured parameters within the knowledge graph and machine learning models. By converting raw data into standardized parameters (geographical entities, concepts, categories, attributes with associated weights and scores), the system improves result relevance while managing data processing complexity through parameter standardization.
3Measurement precision
If multiple tagging levels are applied to search terms, then the interpretation accuracy improves, but the computational resources required increase
Solution Approach 1:
The patent applies partial action by using a weighted finite state transducer that can process queries at multiple tagging levels (geographical entities, concepts, categories, attributes) but only computes and returns the top-scoring interpretations. The machine learning models evaluate multiple possible taggings but selectively process only those that meet threshold criteria, reducing computational energy consumption while maintaining high interpretation accuracy.
Data Source
AI summary
A device implementing a system for providing search results includes at least one processor configured to receive plural terms corresponding to a map-based search query, determine plural interpretations of the map-based search query, each interpretation based on a respective combination of tagging one or more of the plural terms as at least one of a first type, a second type or a third type. The at least one processor is configured to, for each interpretation, determine a set of weight values for the interpretation, based on at least one of context data of the device or a feature of the respective combination, and to assign a score for the interpretation based on the set of weight values for the interpretation. The at least one processor is configured to provide at least one completion suggestion or search result based on the plural interpretations and on the assigned scores.