POI Competition Mining via Graphlet Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing competition relationships between Points of Interest (POIs) rely heavily on expert experience and manual data collection, leading to inaccurate and costly results.
Innovation Solution
A method and apparatus that utilize graphlet mining of user retrieval data to determine occurrence frequencies of preset situations, generating relationship features for POIs, which are then input into a pre-trained prediction model to predict competition relationships, reducing reliance on empirical knowledge and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If expert experience knowledge and manual information collection methods such as questionnaires are used to analyze POI competition relationships, then the analysis can be performed, but the data coverage rate is low and the results deviate greatly from actual competition relationships
Solution Approach 1:
The patent replaces manual information collection methods (questionnaires, expert experience) with automated graphlet mining technology. The system automatically extracts POI relationship features from map retrieval data using graphlet structures and inputs them into a pre-trained prediction model, eliminating the need for manual data collection and significantly improving both data coverage and analysis accuracy.
Solution Approach 2:
The patent creates a computational model (graphlet mining system) that copies and analyzes the natural user retrieval behavior patterns. By mining graphlet structures from actual user map retrieval data, the system captures real competition relationships without manual intervention, producing results that accurately reflect actual user behavior and POI competition dynamics.
2Measurement precision
If expert experience knowledge is used to analyze POI competition relationships, then the analysis can be performed, but the threshold is high and costs increase
Solution Approach 1:
The patent substitutes expert knowledge-based analysis with an automated machine learning system. The pre-trained prediction model processes graphlet mining results automatically, eliminating the need for experts to manually analyze POI relationships. This reduces the knowledge threshold while maintaining or improving analysis accuracy through consistent, scalable automated processing.
Solution Approach 2:
The system performs self-service analysis by automatically mining graphlet features from map retrieval data and feeding them into the prediction model. The entire competition relationship analysis process is autonomous, requiring no expert intervention, thereby reducing both the threshold and cost while maintaining high accuracy through data-driven insights.
3Quantity of substance
If manual information collection methods such as questionnaires are used, then data can be collected, but the data coverage rate is low
Solution Approach 1:
The patent replaces manual questionnaire-based data collection with automated graphlet mining of existing map retrieval data. The system processes large volumes of user retrieval behavior data automatically, achieving high data coverage rates without manual effort. This substitution dramatically improves both data coverage and collection efficiency simultaneously.
Solution Approach 2:
The graphlet mining system serves multiple functions: it extracts POI relationship features, captures user retrieval patterns, and feeds data into the prediction model all through a single automated process. This multi-functional approach efficiently processes diverse data sources to achieve comprehensive data coverage without manual collection efforts.
Data Source
AI summary
A method and apparatus for mining a competition relationship between POIs. An embodiment of the method includes: acquiring a graphlet mining result obtained by mining map retrieval data of users which encompasses attribute information of retrieved target POIs, the graphlet mining result encompassing occurrence frequencies of respective preset situations, and a preset situation comprising: conforming to attribute information of POIs represented by a corresponding preset graphlet and a preset association relationship between attribute information of at least two POIs; for a first and second POI, determining an occurrence frequency of a preset situation corresponding to a preset graphlet where attribute information of the first and second POI co-occur, and generating a relationship feature of the first and second POI; and inputting the relationship feature into a pre-trained relationship prediction model to obtain a competition relationship prediction result of the first POI and the second POI.