Content Recommendation Engine Using User Closeness and Usage Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for recommending electronic contents in social networks fail to accurately reflect the actual use status of contents among users, leading to poor recommendation accuracy, especially when the number of participating users is low.
Innovation Solution
A system that stores closeness information between users and use degree information of electronic contents, using computer processors to select and recommend contents and users based on this data, ensuring more accurate content and user recommendations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If recommendation information is exchanged only between users in an SNS network group, then the system can provide personalized recommendations based on user preferences, but the actual use status of contents by users is not reflected, leading to poor recommendation accuracy
Solution Approach 1:
The system segments the information source into two distinct parts: (1) user preference information exchanged within the SNS network group, and (2) actual use status information collected from a broader population of content users. By separating these information sources and combining them, the system achieves both personalization and accuracy in recommendations.
Solution Approach 2:
The content management server acts as an intermediary that collects use status information from multiple users, processes it, and integrates it with the preference-based recommendations. This intermediary function enables the system to bridge the gap between user preferences and actual content usage patterns, improving recommendation accuracy.
2Measurement precision
If the number of participating users in the SNS network group is insufficient, then the system can maintain user privacy and network quality, but contents are recommended with poor accuracy due to insufficient use status data
Solution Approach 1:
The system merges two different data sources: (1) use status information from a large number of content users (not necessarily SNS group members), and (2) preference information from SNS network group members. This combination allows the system to achieve high recommendation accuracy even when the SNS group size is small, as the use status data comes from a broader population.
Solution Approach 2:
The content management server performs multiple functions: it manages the SNS network group, collects use status information from general users, processes recommendation data, and provides personalized recommendations. This multi-functionality enables the system to overcome the limitation of small SNS group sizes by leveraging universal use status data from the broader user base.
3Adaptability or versatility
If common contents are used by multiple users, then it is likely that the content fits their preferences, but such contents are not always recommended due to lack of use status reflection
Solution Approach 1:
The system implements feedback by continuously collecting use status information from users and using it to adjust and improve recommendations. When common contents are used by multiple users, this use status feedback is reflected in the recommendation algorithm, increasing the reliability of recommending such contents to other users with similar preferences.
Solution Approach 2:
The system changes the parameters used for recommendation by incorporating use status metrics (such as usage frequency, duration, and patterns) into the recommendation algorithm. This parameter change enables the system to identify and recommend common contents that have proven to be suitable for multiple users, improving both adaptability and reliability.
Data Source
AI summary
The system according to an embodiment of the present invention may more appropriately provide recommendation information of contents based on use status of the contents by a plurality of users. The system includes a service provision control unit for controlling provision of services, an information storage unit for storing information, a recommended content selection unit for selecting recommended games based on closeness information between the users and use degree information by the users for each game, a recommended content information sending unit for sending content recommendation information including information specifying recommended games, a recommended user selection unit for selecting recommended users based on the use degree information, a recommended user information sending unit for user recommendation information including information specifying the recommended users, and a relationship request information sending unit for sending relationship request information in response to a request for setting a friend with the recommended users.