User Pattern Matching for Navigation Recommendation Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional car navigation devices struggle to provide accurate operation results when a user's operation history is minimal, leading to suboptimal recommendations.
Innovation Solution
An information processing device with a pattern storage unit, operation result information input unit, pattern specification unit, and operation result selection unit that calculates concordance rates and creates new user patterns based on input operation results, allowing for proper operation result selection even with limited user history.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system relies on user operation history to provide recommendations, then recommendation accuracy improves, but the system cannot provide proper recommendations when user history is minimal
Solution Approach 1:
The system pre-stores multiple user patterns with classification patterns and operation results before the user actually uses the system. These patterns represent typical user behaviors and preferences that are prepared in advance, allowing the system to immediately provide recommendations even when no user history exists yet.
Solution Approach 2:
The system creates simplified copies of user behavior patterns from aggregated data. Instead of requiring individual user history, it copies typical user patterns that have been extracted and stored, allowing new users to receive recommendations based on these representative patterns until they accumulate enough personal history.
2Adaptability or versatility
If the system stores multiple user patterns to handle diverse user behaviors, then recommendation coverage improves, but system complexity increases
Solution Approach 1:
The system segments user behavior into discrete classification patterns (such as preference types, usage habits, or behavioral categories). Each user pattern is divided into multiple classification patterns that can be independently stored and matched, making the complex task of handling diverse user behaviors manageable through structured segmentation.
Solution Approach 2:
The stored user patterns serve multiple functions: they act as templates for matching new user behavior, provide default recommendations for new users, and can be dynamically selected based on current context. This multi-functionality reduces the need for separate mechanisms for different scenarios.
3Measurement precision
If the system calculates concordance rates for all stored patterns, then pattern specification accuracy improves, but processing time increases
Solution Approach 1:
Instead of calculating concordance rates for all possible user patterns equally, the system performs partial calculation by focusing on the most relevant patterns first. It may calculate concordance rates for a subset of patterns that are most likely to match the current user behavior, achieving sufficient accuracy without the computational burden of evaluating every stored pattern.
Data Source
AI summary
The information processing device of the present invention comprises a pattern storage unit which stores a plurality of user patterns containing a classification pattern as a pair of input information that is input based on an input operation related to a predetermined setting item and one among a plurality of operation results that may be selected as a result of the input operation, for each of the plurality of pieces of input information that are mutually different; an operation result information input unit to which is input operation result information related to operation results selected by the user with regard to each of the one or more input operations previously performed by the user; a pattern specification unit which specifies one user pattern among the plurality of stored user patterns based on the input operation result information; and an operation result selection unit which selects one operation result among a plurality of operation results derived from one of the input operations performed by the user based on the specified one user pattern.