Trust Dynamics Clustering for Accurate Human-Agent Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for predicting human trust in automated agents, such as autonomous vehicles, face challenges in quantitatively anticipating trust levels due to limitations in data requirements and individual differences, leading to suboptimal behavior prediction and public acceptance issues.

Innovation Solution

A computer-implemented method for clustering human trust dynamics by receiving and analyzing trust data from participants interacting with agents, identifying phases of interaction, extracting features characterizing trust dynamics, and grouping participants into trust groups based on these features, allowing for customized models that balance data needs and performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If customized trust models are developed for each individual participant, then prediction accuracy is improved, but data requirements increase significantly

Engineering Contradiction:

Improvetrust prediction accuracyVSAvoiddata requirements

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The participant population is segmented into distinct trust dynamics clusters based on behavioral patterns during agent interactions. Instead of creating customized models for every individual, participants are grouped into segments (clusters) that share similar trust characteristics, allowing model reuse across multiple participants while maintaining personalized prediction accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The approach changes the parameter of model customization from individual-level to cluster-level. By identifying key parameters that define trust dynamics and grouping participants with similar parameter profiles, the system achieves personalized prediction without requiring extensive data for each individual, as cluster-level models can be applied to multiple participants.

Inventive Principle:

Principle #35Parameter changes

2Quantity of substance

If a general trust model is used for all participants, then data requirements are reduced, but prediction accuracy decreases due to individual differences

Engineering Contradiction:

Improvedata requirementsVSAvoidtrust prediction accuracy

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

Rather than using a single general model for all participants, the system segments the population into distinct clusters with similar trust dynamics. This allows the use of simplified cluster-level models that require less data than individualized models, while still capturing individual differences through the segmentation approach.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system creates universal cluster-level models that can be applied to multiple participants within each cluster. These multi-functional models serve as generalized representations that work across individuals with similar trust patterns, reducing data requirements while maintaining accuracy through the universality of cluster-based approaches.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If more trust data is collected from each participant, then model personalization is improved, but interaction time and complexity increase

Engineering Contradiction:

Improvemodel personalizationVSAvoiddata collection time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary clustering of participants into trust dynamics groups before detailed model training. By pre-segmenting participants based on initial behavioral patterns, the system establishes personalized model frameworks in advance, reducing the amount of additional data collection needed for each participant while maintaining personalization capabilities.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of collecting extensive data from every participant, the system uses partial data from initial interactions to assign participants to clusters. This partial action approach provides sufficient personalization by leveraging cluster assignments based on limited initial data, avoiding the need for exhaustive data collection while still achieving adaptive modeling.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12145599B2Systems and methods for clustering human trust dynamics

Publication Date: 2024.11.19 HONDA MOTOR CO LTD

AI summary

Systems and methods for clustering human trust dynamics are provided. In one embodiment, a computer implemented method for clustering human trust dynamics is provided. The computer implemented method includes receiving trust data for a plurality of participants interacting with one or more agents in an interaction. The computer implemented method also includes identifying a plurality of phases for the interaction. The computer implemented method further includes extracting features characterizing trust dynamics from the trust data for at least one interaction for each participant of the plurality participants. The at least one interaction is between the participant and an agent of the one or more agents. The computer implemented yet further includes assigning the features characterizing trust dynamics to a phase of the plurality of phases. The computer implemented method includes grouping a subset of the participant of the plurality of participants based on the on features characterizing trust dynamics.