Scholarlink

<p>In academic institutions, researchers usually carry out scientific researches in the form of cooperation, which makes team formation an important issue. Currently, most research on team formation focuses more on communication efficiency, but less on candidate ability, may leading to poor team performance. To address this problem, a quantification model is proposed to quantify the skill level of each candidate. Then, a team formation problem in co-author networks is formalized, which aims at maximizing team skill level and team relationship strength simultaneously. After that, a team formation method based on Non-dominated Sorting Genetic Algorithm is designed. Finally, extensive experiments are conducted on a real dataset to evaluate the performance of the proposed method, and the results demonstrate that the proposed method outperforms the baselines in terms of two optimization objectives, while exhibiting superior stability and convergence speed.</p>

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

Skill Quantification-Based Team Formation Problem for Scientific Research Cooperation

  • 期刊论文

  • 发表日期：

    2025年8月18日

查看全文
  • Zijuan Lu,
  • Kexin Bian,
  • Lichen Zhang,
  • Longjiang Guo,
  • Chenchen Li,
  • Tong Li

摘要

In academic institutions, researchers usually carry out scientific researches in the form of cooperation, which makes team formation an important issue. Currently, most research on team formation focuses more on communication efficiency, but less on candidate ability, may leading to poor team performance. To address this problem, a quantification model is proposed to quantify the skill level of each candidate. Then, a team formation problem in co-author networks is formalized, which aims at maximizing team skill level and team relationship strength simultaneously. After that, a team formation method based on Non-dominated Sorting Genetic Algorithm is designed. Finally, extensive experiments are conducted on a real dataset to evaluate the performance of the proposed method, and the results demonstrate that the proposed method outperforms the baselines in terms of two optimization objectives, while exhibiting superior stability and convergence speed.