Scholarlink

In the dynamic landscape of social media, the efficacy of friend recommendation systems profoundly influences user engagement and interaction quality. This study presents an advanced K-Nearest Neighbors (KNN) model, augmented with refined filters for location, age, and interests, to optimize friend suggestions. Leveraging a comprehensive dataset comprising user IDs, names, dates of birth, specified interests, and precise location details, the model demonstrates remarkable precision in recommending friends based on multifaceted criteria, including country, city, age demographics, and shared interests. The findings underscore a substantial enhancement in recommendation accuracy relative to conventional methodologies, validating the efficiency of the proposed framework. Furthermore, the model showcases adaptability to unlabeled data, exhibiting resilience in capturing user preferences and adjusting to evolving social dynamics. By prioritizing precision and relevance, the model steers friend recommendations towards unparalleled alignment with intricate user preferences, facilitating more profound and meaningful digital connections in the ever-evolving social media landscape.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

K-Nearest Neighbor Based Friends Suggestion on Social Media

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yash Vadukiya,
  • Shraddha Parmar,
  • Vaishali Vadhavana,
  • Krishna Patel,
  • Bansari Patel

摘要

In the dynamic landscape of social media, the efficacy of friend recommendation systems profoundly influences user engagement and interaction quality. This study presents an advanced K-Nearest Neighbors (KNN) model, augmented with refined filters for location, age, and interests, to optimize friend suggestions. Leveraging a comprehensive dataset comprising user IDs, names, dates of birth, specified interests, and precise location details, the model demonstrates remarkable precision in recommending friends based on multifaceted criteria, including country, city, age demographics, and shared interests. The findings underscore a substantial enhancement in recommendation accuracy relative to conventional methodologies, validating the efficiency of the proposed framework. Furthermore, the model showcases adaptability to unlabeled data, exhibiting resilience in capturing user preferences and adjusting to evolving social dynamics. By prioritizing precision and relevance, the model steers friend recommendations towards unparalleled alignment with intricate user preferences, facilitating more profound and meaningful digital connections in the ever-evolving social media landscape.