Identification of Potential Cooperation Relationships Among Scientists

Area of Science:

• Network Science
• Bibliometrics
• Computational Social Science

Background:

• Scientist cooperation networks are crucial for research advancement.
• Understanding information spread dynamics is key to network evolution.
• Link prediction aids in identifying potential collaborations.

Purpose of the Study:

• To analyze information spread in scientist cooperation networks.
• To develop an improved link prediction algorithm for these networks.
• To identify potential collaboration opportunities for scientists.

Main Methods:

• Abstracting real networks into simulated networks.
• Simulating information spread using an improved SIS model.
• Developing a hybrid weighted link prediction algorithm incorporating node attributes and spread factors.

Main Results:

• The improved SIS model effectively simulates information spread in cooperation networks.
• The hybrid weighted link prediction algorithm significantly improves prediction accuracy.
• Experimental results validate the propagation model and link prediction algorithm on both simulated and real networks.

Conclusions:

• Information spread analysis reveals network formation laws.
• Link prediction methods preserve network information integrity.
• The hybrid algorithm offers practical suggestions for scientists seeking partners, enriching cooperation networks.