Community Detection via Shared Entity Access Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current approaches for detecting and generating communities of users in distributed storage systems focus solely on connections between nodes, neglecting the actual attributes common to the nodes.
Innovation Solution
A community detection system that identifies communities based on a common set of entities accessed by users in a distributed storage system, using techniques such as selecting an anchor entity with the greatest number of distinct users and forming communities around shared entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current community detection approaches focus only on connections between nodes, then the detection process is simple, but the communities formed do not reflect actual common attributes of users
Solution Approach 1:
The patent transitions from analyzing only connection-based relationships (single dimension) to incorporating both connection information and entity access pattern attributes (multiple dimensions). By adding the attribute dimension of common entities accessed by users, the system achieves more accurate community detection that reflects actual user similarities beyond just network connections.
2Stability of the object's composition
If communities are formed based on connection patterns only, then the formation process is computationally efficient, but the communities lack cohesion based on actual shared attributes
Solution Approach 1:
The system pre-processes and stores entity access patterns for each user before community detection is performed. This preliminary action of organizing access pattern data allows the actual community formation process to efficiently compare pre-computed attributes rather than analyzing raw access logs in real-time, thus maintaining speed while improving cohesion.
3Reliability
If entity access patterns are incorporated into community detection, then communities better represent actual user groups, but the detection process becomes more complex
Solution Approach 1:
The patent extracts entity access patterns as separate, identifiable attributes from user behavior data. By isolating and structuring these patterns into distinct data elements that can be independently compared, the system incorporates rich attribute information without requiring overly complex detection algorithms, thus improving reliability while managing complexity.
Data Source
AI summary
A system, method, and machine-readable storage medium for forming a community based on a common set of attributes are provided. In some embodiments, the method includes creating a list of entities associated with a plurality of users, each entity included in the list of entities being accessed by a user of the plurality of users. The method also includes identifying a first entity accessed by a group of users of the plurality of users. The method further includes determining a first set of entities accessed by each user of the group of users, the first set of entities being included in the list of entities. The method also includes removing the first entity and the first set of entities from the list of entities. The method further includes forming a first community including the group of users, the first entity, and the first set of entities.