Advertising Recommendation Algorithm Based on User's Social Network

With the popularity of social media platforms, social networks have become an important bridge connecting users and businesses. In order to recommend ads to users more accurately, this study proposes a novel AD recommendation algorithm, which integrates the advantages of the four original methods to improve the relevance and user experience of AD recommendation. Content-based Filtering: Models users' interests by analyzing their past behavior and preferences, such as likes, comments, and types of Content shared. New algorithms use this information to predict what ads are likely to be of interest to users, enabling personalized recommendations. Collaborative Filtering: Learn from the method of similarity calculation among users to find out user groups with similar interests or behavior patterns. The new approach not only considers direct friend relationships, but also explores indirect connections in multi-tiered social network structures to broaden the range of potential target audiences. Social Graph Weighting: introduces the concept of the strength of relationships between nodes in a social network and assigns different weights based on interaction frequency and intimacy. This allows algorithms to more accurately assess social impact and prioritize goods and services that are associated with close contacts or influencers. Context-awareness: Adjust recommendation strategies based on real-time factors such as time and place. Considering the different needs of users in different situations, the new algorithm dynamically captures these changes to ensure that the most relevant product or service information is provided at the right time.