Social Network Friend Recommendation Based on Intimacy Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional social network service (SNS) systems fail to effectively recommend friends of friends by solely relying on mutual friend counts, neglecting the actual intimacy and relationship between users, leading to irrelevant friend recommendations.
Innovation Solution
A system and method that extract and calculate communication information, such as interaction frequency and relationship duration, to determine intimacy levels between users, and generate a list of recommended friends based on these calculated intimacy values, prioritizing closer relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If friend recommendation is based solely on mutual friend count, then the recommendation process is simple and fast, but the recommendation accuracy and relevance deteriorate
Solution Approach 1:
The patent transforms the single parameter of mutual friend count into multiple parameters including communication frequency, communication time, and interaction intensity. These parameters are combined through a weighted calculation model to produce an intimacy score, thereby improving measurement precision while maintaining processing efficiency through systematic parameter integration.
Solution Approach 2:
The patent adds new dimensions to the friend recommendation system by incorporating temporal dimensions (communication time), frequency dimensions (interaction count), and intensity dimensions (communication depth). This multi-dimensional approach enriches the measurement of friend relationships beyond the traditional single-dimensional mutual friend count.
2Measurement precision
If multiple communication factors are extracted and calculated to determine intimacy, then friend recommendation accuracy improves, but system complexity increases
Solution Approach 1:
The patent segments the complex intimacy calculation into distinct modular components: communication frequency extraction, communication time extraction, interaction intensity calculation, and weighted integration. Each component operates independently and can be adjusted separately, reducing overall system complexity while maintaining high measurement precision.
Solution Approach 2:
The patent creates a universal intimacy calculation framework that can be applied across different social network contexts and user relationships. The same set of parameters and calculation methods serves multiple purposes including friend recommendation, relationship analysis, and network optimization, thereby managing complexity through functional consolidation.
3Ease of manufacture
If conventional mutual friend count method is used, then the recommendation system is easy to implement, but it fails to reflect actual user relationships
Solution Approach 1:
The patent introduces communication patterns as an intermediary layer between the simple mutual friend count and the complex reality of user relationships. By measuring and analyzing communication behaviors (frequency, time, intensity), the system creates a reliable proxy that reflects actual relationship dynamics without requiring direct observation of all relationship aspects.
Data Source
AI summary
A social network service (SNS) system and method for recommending a friend of a friend based on an intimacy between users are provided. The SNS system includes an extracting unit to extract communication information from each of a friend relationship between a user and a first-group friend and a friend relationship between the user and a second-group friend. The system also includes a calculating unit to calculate an intimacy of each of the friend relationships based on the extracted communication information. The system includes a generating unit to generate a list of recommended second-group friends based on the calculated intimacy.