Scholarlink

Campus placement is a crucial process for universities to find and recruit bright students for preliminary employment and internships. An institution’s reputation and yearly enrollment are inevitably linked to its capacity to place students. As a result, most colleges strive extensively to bolster their placement sections in order to raise their overall reputation. Assistance in this specific field can make a significant impact on the college’s ability to place its students successfully. This study predicts student placement outcomes using Bayesian Belief Networks (BBNs) by utilizing Hill Climb Search and Tree Search algorithms on over 10,000 student records. Hill Climb Search achieved 86.022% accuracy versus 74.6% for Tree Search, with superior performance across multiple evaluation metrics. The study also integrates a Llama 2 RAG model for resume-job matching, enhancing practical applicability. This combined approach of BBNs and advanced language models offers valuable insights for improving campus placement processes and outcomes.

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

Optimizing Uncertainty in Placement Prediction Using Bayesian Belief Networks and LLMs

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Aarya Kamdar,
  • Bhuvi Ghosh,
  • Kanchan Dabre

摘要

Campus placement is a crucial process for universities to find and recruit bright students for preliminary employment and internships. An institution’s reputation and yearly enrollment are inevitably linked to its capacity to place students. As a result, most colleges strive extensively to bolster their placement sections in order to raise their overall reputation. Assistance in this specific field can make a significant impact on the college’s ability to place its students successfully. This study predicts student placement outcomes using Bayesian Belief Networks (BBNs) by utilizing Hill Climb Search and Tree Search algorithms on over 10,000 student records. Hill Climb Search achieved 86.022% accuracy versus 74.6% for Tree Search, with superior performance across multiple evaluation metrics. The study also integrates a Llama 2 RAG model for resume-job matching, enhancing practical applicability. This combined approach of BBNs and advanced language models offers valuable insights for improving campus placement processes and outcomes.