Exploiting Publication Contents and Collaboration Networks for Collaborator Recommendation

Area of Science:

• Computer Science
• Information Science
• Bibliometrics

Background:

• Online social networks facilitate researcher communication and collaboration.
• Identifying relevant and potential collaborators remains a significant challenge in scientific communities.

Purpose of the Study:

• To propose a novel collaborator recommendation model, CCRec.
• To enhance the accuracy and effectiveness of identifying potential research collaborators.

Main Methods:

• CCRec integrates researchers' publication data and collaboration network information.
• A topic clustering model is employed to identify academic domains.
• A random walk model is utilized to compute researchers' feature vectors.

Main Results:

• Benchmarking experiments were conducted using DBLP datasets.
• CCRec demonstrated superior performance compared to existing state-of-the-art methods.
• The model achieved higher precision, recall, and F1 scores.

Conclusions:

• CCRec effectively identifies potential research collaborators by combining publication and network data.
• The proposed model offers a significant improvement over current collaborator recommendation techniques.