Standardizing Device Metrics for Automated Assistant Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Comparing the behavior of client devices with different software and hardware configurations is challenging due to the variability in device-specific metrics, making it difficult to generate aggregate metrics and analyze the behavior of automated assistants across multiple devices.

Innovation Solution

The implementation involves generating standardized metrics from device-specific metrics using conversion mappings based on device characteristics, allowing for comparison and analysis of device behavior in a standardized manner.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If device-specific metrics are used directly for analysis, then device-specific behavior can be monitored, but comparing behaviors across disparate devices becomes difficult and requires multiple analysis programs

Engineering Contradiction:

Improvedevice behavior monitoring accuracyVSAvoidanalysis program complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms device-specific metrics into standardized metrics by applying conversion mappings that adjust timing values based on device characteristics. This parameter transformation enables uniform analysis across different devices while preserving the ability to account for device-specific variations through the conversion process

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent creates a universal analysis program that can process metrics from any device type by standardizing the metric format. The single analysis program replaces multiple device-specific programs, achieving multi-functionality while maintaining accurate behavior monitoring through the standardized metric structure

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Adaptability or versatility

If multiple analysis programs are executed simultaneously on remote servers to handle different device metrics, then comprehensive device coverage is achieved, but significant memory and computational resources are consumed

Engineering Contradiction:

Improvedevice compatibilityVSAvoidremote server resource consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent merges multiple device-specific analysis programs into a single universal analysis program that processes standardized metrics. This consolidation reduces remote server resource consumption by eliminating the need to run and maintain multiple separate programs while preserving the ability to analyze metrics from diverse device types

Inventive Principle:

Principle #5Merging (Combining)

3Adaptability or versatility

If multiple analysis programs are maintained and updated separately, then device-specific analysis requirements are met, but significant programming effort and client device resources are required

Engineering Contradiction:

Improvedevice-specific analysis capabilityVSAvoidprogram maintenance effort

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent develops a universal analysis program that handles multiple device types through standardized metric processing. This single program eliminates the need to separately maintain and update multiple device-specific programs, significantly reducing programming effort and client device resource requirements while maintaining comprehensive device coverage

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Device complexity

If device metrics are converted to standardized metrics using conversion mappings, then a single analysis program can be used, but additional processing steps are required

Engineering Contradiction:

Improveanalysis program simplicityVSAvoidmetric processing efficiency

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent performs the conversion from device-specific metrics to standardized metrics as a preliminary step before analysis. By pre-processing the metrics into a standardized format using conversion mappings, the system enables the use of a single simplified analysis program while the conversion overhead is managed through efficient mapping algorithms

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12316517B2Standardizing analysis metrics across multiple devices

Publication Date: 2025.05.27 GOOGLE LLC

AI summary

Implementations relate to generating standardized metrics from device specific metrics that are generated during an interaction between a user and an automated assistant. The metrics indicate events that occurred while processing an interaction of a user with the automated assistant and are specific to the particular configuration of the device with which the user is interacting. Conversion mappings are determined based on device characteristics that can be utilized to convert the device metrics into standardized metrics. Analysis metrics are generated based on the standardized metrics that are incapable of being generated from the device metrics. Some implementations include visually rendering the analysis metrics such that one or more of the analysis metrics are rendered more prominently than other metrics.