Context-Aware Destination Recommendation System Using Entropy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Generic mapping and navigation applications fail to provide personalized point-of-interest recommendations tailored to users' desires and needs, lacking context awareness which results in irrelevant or annoying suggestions.

Innovation Solution

A system that determines user context and mobility patterns to predict destination probabilities, using entropy information to decide when to present personalized recommendations, avoiding disturbance during routine activities and focusing on unfamiliar or relevant locations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If generic mapping applications provide general point of interest recommendations, then coverage and availability of recommendations is improved, but relevance to user's desires and needs deteriorates

Engineering Contradiction:

Improverecommendation coverageVSAvoidrecommendation relevance

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by customizing recommendations based on user-specific context factors such as mobility patterns, preferences, and current situation. Instead of uniform generic recommendations, the system adapts the content and delivery of recommendations to match individual user characteristics, thereby improving relevance while maintaining broad coverage through context-aware filtering and selection

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes parameters by dynamically adjusting recommendation delivery based on context variables including user location, time, mobility behavior, and expressed preferences. By modifying recommendation parameters (what to recommend, when to recommend, and how to present) based on changing context, the system achieves both broad adaptability and precise relevance

Inventive Principle:

Principle #35Parameter changes

2Productivity

If recommendations are provided continuously, then user engagement opportunities are improved, but user disturbance and annoyance increases

Engineering Contradiction:

Improveengagement opportunitiesVSAvoiduser disturbance

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The system implements periodic action by delivering recommendations at strategically determined intervals rather than continuously. It monitors user context and mobility patterns to identify optimal moments when users are most receptive, creating periodic engagement opportunities that avoid overwhelming the user while maintaining meaningful interaction rates

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system uses feedback mechanisms to monitor user responses to recommendations and adjust future delivery accordingly. By tracking whether users accept, ignore, or express negative reactions to recommendations, the system learns optimal timing and frequency, reducing disturbance while preserving engagement opportunities through adaptive delivery schedules

Inventive Principle:

Principle #23Feedback

3Measurement precision

If context analysis and entropy calculation are implemented, then recommendation precision is improved, but system complexity increases

Engineering Contradiction:

Improvedestination prediction accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system applies partial action by calculating entropy and analyzing context factors selectively rather than comprehensively for all users and all times. It focuses computational resources on users and situations where context analysis will most improve recommendation value, performing simplified entropy calculations only when mobility patterns indicate uncertainty about user intentions, thereby reducing overall complexity while maintaining precision where needed

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS9648463B2Method and apparatus for providing context-related point-of-interest recommendations

Publication Date: 2017.05.09 HERE GLOBAL BV

AI summary

An approach is provided for providing context-related destination location recommendations. The destination recommendation platform determines at least one context associated with at least one user. The platform processes and/or facilitates a processing of mobility pattern information associated with the at least one user to determine one or more predicted destination locations and associated probability information based, at least in part, on the at least one context. The probability information represents respective one or more probabilities that the at least one user will travel to the one or more predicted destination locations under the at least one context. The platform further determines entropy information based, at least in part, on the probability information. The platform determines whether to cause, at least in part, a presentation of one or more recommended destination locations to the at least one user based, at least in part, on the entropy information.