Graph Database POI Recommendation Using User Intimacy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current POI recommendation systems face inefficiencies due to pre-calculating all possible POIs, leading to resource waste, and lack personalized recommendations due to unsuitable data storage formats for graph traversal, which prevents the use of real-time position and relationship network data.
Innovation Solution
A POI recommendation method and device based on a graph database that acquires a feature vector and friend intimacy coefficient of a target user, and a feature matrix of related POI consumption records to determine personalized POIs by calculating similarities and querying relevant data from a graph database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If offline calculation and preparation is performed for all possible POIs in advance, then the recommendation system can provide comprehensive POI coverage, but it causes great waste of calculation resources
Solution Approach 1:
The patent extracts and utilizes only the necessary subset of POI data for each user based on their real-time location and historical behavior, rather than processing all possible POIs. The graph database enables selective traversal to retrieve only relevant POI information needed for personalized recommendations, eliminating computation on irrelevant data.
Solution Approach 2:
The system performs preliminary actions by pre-building the graph database structure with user relationships, POI data, and consumption records during offline preparation. This allows the online recommendation phase to efficiently query pre-processed data without performing heavy calculations in real-time, reducing online computational resource waste.
2Quantity of substance
If traditional data warehouse storage format is used, then the system can store large amounts of data, but it is not suitable for graph traversal to acquire data for personalized recommendations
Solution Approach 1:
The patent changes the storage parameter from traditional relational data warehouse format to graph database format. This transformation enables efficient graph traversal operations while maintaining the ability to store large amounts of data including user relationships, POI information, and consumption records in a structure optimized for recommendation queries.
3Device complexity
If real-time position and relationship network data are not utilized, then the system simplifies data processing, but it leads to lack of personality in recommendation results
Solution Approach 1:
The patent segments the recommendation process into distinct components: real-time position data processing, relationship network traversal, historical consumption analysis, and POI filtering. Each segment handles specific data types and operations, enabling the system to process multiple data sources efficiently while delivering personalized recommendations that adapt to individual user preferences and contexts.
Data Source
AI summary
A POI recommendation method based on the graph database includes: acquiring an ID and positioning information of a target user; according to the ID and the positioning information of the target user, acquiring a feature vector of the target user, a friend intimacy coefficient of the target user and a target POI set from a target graph database, wherein the target POI set is a set composed of POIs within a preset distance range queried according to the positioning information; according to the target POI set, acquiring a feature matrix of a user set with related POI consumption records from the target graph database, wherein the related POI consumption records are consumption records associated with any POI in the target POI set; and according to the friend intimacy coefficient, the feature vector of the target user and the feature matrix, determining a target POI from the target POI set.