Geographic Search Query Interpretation via Template Probability

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing geographic search query systems face difficulties in interpreting queries with varying structures and languages, leading to incorrect and inaccurate responses due to differences in user inputs from diverse countries and languages.

Innovation Solution

A computer-implemented method and system that generates query templates from logs of executed geographical search queries, determines probability distributions for these templates based on context, and provides interpretation candidates for incoming queries, improving accuracy by ranking potential interpretations based on probability and user behavior.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If geographic search queries are interpreted using simple matching methods, then the system is easy to operate, but the interpretation accuracy deteriorates due to varying query structures and languages

Engineering Contradiction:

Improveease of query interpretationVSAvoidquery interpretation accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces query templates as intermediary structures that mediate between raw user queries and interpretation results. These templates serve as a standardized framework that accommodates various query structures and languages while maintaining consistent interpretation accuracy. The templates act as a bridge that transforms diverse input formats into uniform interpretation candidates.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically adjusts interpretation parameters based on the probability distribution associated with different query templates. By changing the parameters of query interpretation (such as term type assignments and structural patterns) according to learned probability distributions, the system adapts to varying query structures and languages while maintaining high accuracy.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If the system uses multiple query templates and probability distributions to improve interpretation accuracy, then the interpretation precision improves, but the system complexity increases

Engineering Contradiction:

Improvequery interpretation accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by pre-generating query templates and their associated probability distributions from historical query logs before actual query interpretation occurs. This offline preparation work creates a ready-to-use framework of templates that can be quickly applied during runtime, reducing the computational complexity of real-time interpretation while maintaining high accuracy through pre-computed probability distributions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates simplified copies of query patterns through templates that represent common query structures. Instead of processing each unique query from scratch, the system uses template copies that capture the essential structure and semantics of frequently occurring query types, reducing system complexity while preserving interpretation accuracy through probabilistic matching.

Inventive Principle:

Principle #26Copying

3Measurement precision

If users enter longer and more specific queries to improve accuracy, then the interpretation accuracy improves, but the query length increases

Engineering Contradiction:

Improveinterpretation accuracyVSAvoidquery length

Core Design Contradiction:

Measurement precisionVSLength of moving object

Solution Approach 1:

The query templates provide a universal framework that can handle multiple query types and structures with a single interpretation mechanism. This multi-functionality allows the system to accurately interpret both short and long queries using the same template-based approach, eliminating the need for users to extend query length to achieve accurate interpretation.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system replaces the mechanical approach of requiring longer queries with a probabilistic template-matching system. Instead of relying on query length to disambiguate meanings, the system uses learned probability distributions from historical data to determine the most likely interpretation, allowing short queries to achieve the same accuracy as traditionally required longer queries.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10127245B2Systems, methods, and computer-readable media for interpreting geographical search queries

Publication Date: 2018.11.13 GOOGLE LLC

AI summary

Systems, methods, and computer-readable media for interpreting geographical search queries are provided. Query templates for geographical search queries are generated from search logs of previously executed geographical search queries. Unique queries are extracted from the search logs and interpretations for these queries are obtained. Based on these interpretations, query templates having an arrangement of term types are extracted. Additionally, probability distributions for the query templates for a context, such as a locale, language, client type, etc., are determined.