Machine Learning Module for Augmented Reality Locale Attraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

There is a need for improved techniques to enhance user experiences by providing more effective deployment of augmented reality representations in augmented reality displays, particularly in suggesting locales of interest to users based on their current location and available time.

Innovation Solution

A machine learning module is used to determine locale attraction scores by combining user features and locale characteristics, which are then used to render augmented reality representations of qualifying locales in an AR display, guiding users to likely interest areas and increasing the likelihood of positive experiences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If augmented reality representations are provided for all possible locales, then user information completeness is improved, but device complexity and processing requirements increase

Engineering Contradiction:

Improveuser information completenessVSAvoidprocessing complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent applies local quality by providing augmented reality representations selectively for specific locales rather than all locales. The system determines locale attraction scores and provides AR representations only for locales exceeding a threshold, making the information delivery targeted and efficient while maintaining completeness for relevant locations.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes parameters by calculating locale attraction scores based on multiple factors (user preferences, locale characteristics, temporal/spatial context) and using these scores to dynamically determine which locales receive AR representations. This parameter-based filtering optimizes the balance between information completeness and processing complexity.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If locale attraction scores are calculated for all locales, then recommendation accuracy is improved, but loss of time increases

Engineering Contradiction:

Improverecommendation accuracyVSAvoidcalculation time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the necessary information for calculating locale attraction scores by selectively gathering user feature values and locale feature values relevant to the current context. The system determines qualifying locales based on available time and location, then calculates scores only for these filtered locales rather than all possible locales, reducing calculation time while maintaining accuracy.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary actions by pre-determining qualifying locales based on user available time and current location before calculating attraction scores. This preliminary filtering reduces the number of locales requiring detailed score calculations, thereby reducing overall computation time while preserving recommendation accuracy for the most relevant locations.

Inventive Principle:

Principle #10Preliminary action

3Productivity

If augmented reality representations are provided for more locales, then user engagement is improved, but loss of time for processing and transmission increases

Engineering Contradiction:

Improveuser engagementVSAvoidprocessing and transmission time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system optimizes processing and transmission time by changing parameters dynamically - it calculates locale attraction scores based on user features and locale features, then transmits information only for locales exceeding a threshold score. This parameter-based selection ensures high user engagement through personalized recommendations while minimizing processing and transmission time by excluding low-relevance locales.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11789685B1Training and using a machine learning module to determine locales and augmented reality representations of information on locales to render in an augmented reality display

Publication Date: 2023.10.17 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Provided are a computer program product, system, and method for training and using a machine learning module to determine locales and augmented reality representations of information on locales to render in an augmented reality display. A determination is made of qualifying locales accessible to a visiting user during a user available time from a current location. Values for locale features of the qualifying locales and values of user features in a user profile of the visiting user are provided as input to a locale attraction machine learning module to determine local attraction scores for the qualifying locales. A determination is made of qualifying locales based on locale attraction scores of the qualifying locales to transmit to the visiting user computing device to cause an augmented reality display to render augmented reality representations of information on the qualifying locales.