On-Device Query Completion for Entity Search

Solution Overview

Problem

Users face difficulties in accessing specific entities on client devices due to the need to accurately identify and launch applications, leading to repeated attempts and resource consumption when queries are incomplete or inadequate.

Innovation Solution

An on-device system identifies associated applications based on historical queries to automatically complete user queries, suggesting the correct application for accessing entities, reducing the need for multiple inputs and minimizing resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If users provide incomplete or inadequate queries to access entities, then the automated assistant may not be responsive and requires repeated attempts, but this leads to extensive consumption of computing and battery resources

Engineering Contradiction:

ImproveQuery input simplicityVSAvoidBattery resource consumption

Core Design Contradiction:

Ease of operationVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary action by analyzing historical queries to pre-establish associations between entities and applications before the user makes their query. When a user provides an incomplete query, the system has already prepared the mapping information needed to automatically complete the query, eliminating the need for repeated attempts and reducing energy consumption.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The automated assistant performs self-service by automatically completing incomplete queries using historical data analysis. The system autonomously determines the intended application based on entity associations from historical queries, eliminating the need for users to provide complete queries manually and reducing the iterative energy consumption that would otherwise occur.

Inventive Principle:

Principle #25Self-service

2Ease of operation

If users provide incomplete or inadequate queries to access entities, then the automated assistant may not be responsive and requires repeated attempts, but this leads to repeated attempts by the user

Engineering Contradiction:

ImproveQuery input simplicityVSAvoidTime for repeated query attempts

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The system performs preliminary action by analyzing historical queries to pre-establish associations between entities and applications before the user makes their query. When a user provides an incomplete query, the system has already prepared the mapping information needed to automatically complete the query, eliminating the need for repeated attempts and reducing energy consumption.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The automated assistant performs self-service by automatically completing incomplete queries using historical data analysis. The system autonomously determines the intended application based on entity associations from historical queries, eliminating the need for users to provide complete queries manually and reducing the iterative energy consumption that would otherwise occur.

Inventive Principle:

Principle #25Self-service

3Reliability

If the system requires accurate identification of applications to trigger them, then applications can be precisely activated, but users must accurately identify the application name which increases query complexity

Engineering Contradiction:

ImproveApplication activation accuracyVSAvoidQuery specification requirement

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system introduces an intermediary layer of automatic query completion that mediates between the user's simple entity reference and the required application-specific action. The historical query analysis system acts as a mediator that translates incomplete user queries into complete application-triggering queries automatically, maintaining activation accuracy while reducing query complexity requirements.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The automated assistant performs self-service by automatically completing incomplete queries using historical data analysis. The system autonomously determines the intended application based on entity associations from historical queries, eliminating the need for users to provide complete queries manually and reducing the iterative energy consumption that would otherwise occur.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20240411813A1Rendering suggestion for searching entity within application in response to determining the entity is associated with the application

Publication Date: 2024.12.12 GOOGLE LLC

AI summary

Implementations include determining that an entity is associated with an application, determining fulfillment information for fulfilling a search of the entity within the application, and, in response to determining that the entity is associated with the application, storing, at a computing device, an association between the entity and the fulfillment information for fulfilling the search of the entity within the application. Some of those implementations further include, subsequent to the storing, receiving, via the computing device, incomplete user input that indicates the entity but not the application, and in response to receiving the incomplete user input, causing a suggestion to be rendered via the computing device in a selectable format based on the stored association between the entity and the fulfillment information. The suggestion in the selectable format, when selected, executes the fulfillment information to fulfill the search of the entity within the application.