Lexicographic Optimization for Context-Aware POI Recommendations

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current point-of-interest (POI) recommendation systems fail to accurately and efficiently recommend POIs based on a user's current context, intent, and emotions, often relying on historical data and being prone to errors, especially in cases of limited user interaction history or privacy concerns, and are not robust to changes in user behavior.

Innovation Solution

A computer-implemented method using lexicographic optimization with approximation ratios to select and rate POIs based on user-defined current state functions, incorporating utility measures and approximation ratios to provide secure and robust recommendations that adapt to user preferences without requiring personal data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If POI recommendation systems rely on historical user data and traditional algorithms, then they can provide personalized recommendations, but they fail to accurately capture user current context, intent, and emotions, and are prone to errors when user interaction history is limited

Engineering Contradiction:

Improveaccuracy of capturing user context and intentVSAvoidrobustness to changes in user behavior

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent transforms the recommendation approach by changing from relying on historical behavior patterns to using real-time contextual parameters. Users explicitly provide current state information (intent, emotions, context) which serves as new parameters for the recommendation function, replacing the traditional reliance on historical data patterns.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces an intermediary layer between user behavior and recommendations: the current state function. This intermediary captures user intent, emotions, and context explicitly, serving as a mediator that translates user needs into POI recommendations without relying on historical behavior inference.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If POI recommendation systems use explicit user input to capture current context, then they can improve accuracy, but they require more user interaction and input effort

Engineering Contradiction:

Improveaccuracy of POI recommendationsVSAvoiduser input effort

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent applies partial action by requiring users to provide only essential current state information (intent, emotions, context) rather than comprehensive historical data. This partial input approach achieves accurate recommendations without demanding excessive user interaction or historical data provision.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If POI recommendation systems prioritize personalization based on historical behavior, then they can adapt to individual users, but they are prone to errors in cases of limited user interaction history or privacy concerns

Engineering Contradiction:

Improveadaptation to individual user preferencesVSAvoidaccuracy with limited user history

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent inverts the traditional personalization approach: instead of inferring user preferences from historical behavior patterns, the system directly receives explicit user input about current state (intent, emotions, context). This inversion allows personalization without relying on historical data, solving the problem of limited user history and privacy concerns.

Inventive Principle:

Principle #13The other way round (Inversion)

4Productivity

If POI recommendation systems use traditional algorithms based on historical data, then they can provide recommendations, but they are not robust to changes in user behavior and context

Engineering Contradiction:

Improverecommendation generation capabilityVSAvoidrobustness to behavior changes

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent makes the recommendation system dynamic by using real-time current state functions that adapt to changing user intent, emotions, and context. Unlike static historical pattern matching, the system dynamically adjusts recommendations based on the user's current situation, ensuring robustness to behavior changes.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS11775604B2Method of locating points of interest in a geographic area

Publication Date: 2023.10.03 NAVER CORP

AI summary

A computer-implemented method of locating points of interest for a user in a geographic area is disclosed. A current state of the user is used as an index into a list of a plurality of current state functions to select at least one of the plurality of current state functions. Each current state function corresponds with a utility measure set where each utility measure set includes a plurality of utility measures ordered in a selected sequence. The computer is used to perform: (i) lexicographic optimization with respect to the plurality of current state functions so that the selected sequence of each utility measure set is optimally ordered, and (ii) rate a plurality of points of interest with respect to at least one of the plurality of optimized current state functions, and, based on the rating, to automatically select a set of points of interest for the user.