Peer Review #1 of "Dual network embedding for representing research interests in the link prediction problem on co-authorship networks (v0.2)"

Ilya Makarov, Olga Gerasimova, Pavel Sulimov, Leonid Zhukov, S Abu-El-Haija, B Perozzi, R Al-Rfou, S Adafre, M De Rijke, L Backstrom, J Leskovec, M Belkin, P Niyogi, H Cai, V Zheng, K Chang, S Cao, W Lu, Q Xu, N Cetorelli, S Peristiani, S Chang, W Han, J Tang, G.-j Qi, C Aggarwal, T Huang, H Chen, X Li, Z Huang, H Cho, Y Yu, P Cui, X Wang, J Pei, W Zhu, F Gao, K Musial, C Cooper, S Tsoka, S Gao, L Denoyer, P Gallinari, P Goyal, E Ferrara, A Grover, Q He, D Kifer, P Mitra, L Giles, X Huang, J Li, X Hu, X Kong, M Mao, W Wang, J Liu, B Xu, G Kossinets, D Watts, Q Le, T Mikolov, Y Liang, Q Li, T Qian, D Liben-Nowell, J Kleinberg, Y Liu, Z Kou, L L, T Zhou, O Bulanov, N Meshcheryakova, I Karpov, K Korovina, M Mcpherson, L Smith-Lovin, J CookOpenalex Unknown 2019 · DOI · OpenAlex

Abstract

We present a study on co-authorship network representation based on network embedding together with additional information on topic modeling of research papers and a new edge embedding operator. We use a link prediction model for constructing a recommender system for searching collaborators with similar research interests. By extracting topics for each paper, we construct a keyword co-occurrence network and use its embedding for further generalizing author attributes. Standard graph feature engineering and network embedding methods were combined for constructing a co-author recommender system formulated as a link prediction problem and prediction of future graph structure. We evaluate our survey on a dataset containing temporal information on National Research University Higher School of Economics over twenty-five years of research articles indexed in Russian Science Citation Index and Scopus. Our model of network representation shows better performance for stated binary classification tasks on several co-authorship networks.

Introduction

In the rapidly expanding landscape of academic publishing, identifying suitable collaborators has become a significant challenge. Traditional recommender systems often rely on simplistic metrics like citation counts or centrality, which fail to capture the nuanced structural identity of researchers or the evolving trajectory of their interests. This paper addresses this gap by framing collaboration discovery as a link prediction problem: predicting whether two authors will collaborate in the future based on current network snapshots and research profiles.

Methodology

The researchers propose a dual-embedding framework that integrates topological data with semantic research interests:

  1. Dual Network Construction: Two distinct networks are built—a standard co-authorship network (where nodes represent authors) and a keyword co-occurrence network (where nodes represent research topics).
  2. Embedding with node2vec: The node2vec algorithm transforms nodes from both networks into low-dimensional vectors, capturing the local and global context of authors and keywords based on their neighborhood in the graph.
  3. Neighbor Weighted-L2 Operator: To represent potential edges between authors, the authors introduced a novel operator that incorporates information from the nodes' neighbors. This captures local community structure more effectively than standard operators like the Hadamard product or simple averaging.
Visualization of the co-authorship network structure, showing global connectivity (A) and detailed local clusters (B, C, D) that inform the embedding process.

Experimental Results

The model was evaluated across three datasets: a real-world dataset from the National Research University Higher School of Economics (NRU HSE), a curated Scopus subset, and the large-scale AMiner benchmark to test scalability.

Key findings include:

  • Superior Edge Representation: The Neighbor Weighted-L2 operator consistently outperformed standard baselines across almost all metrics, suggesting that a researcher's colleagues' connections are strong predictors of their own future links.
  • Predictive Accuracy: On the Scopus dataset, the combined model (topology + research interests) achieved a ROC-AUC of 0.878 and an Accuracy of 0.866.
  • Feature Synergy: Integrating research interest embeddings with structural embeddings provided a significant performance boost compared to using either feature set in isolation, confirming that "Dual" is better.

Discussion and Conclusion

While the dual network approach shows great promise, challenges remain regarding data sparsity in networks with few publications and the computational complexity of embedding large-scale graphs. The authors found that Support Vector Machines (SVM) performed most consistently on the sparse, high-dimensional feature spaces generated. Future work will explore deep learning solutions, such as graph auto-encoders, to learn node and edge representations simultaneously and capture more complex non-linear relationships within academic networks.

Summary updated

Abstract

We present a study on co-authorship network representation based on network embedding together with additional information on topic modeling of research papers and new edge embedding operator.We use link prediction model for constructing recommender system for searching collaborators with similar research interests.Extracting topics for each paper we construct keywords co-occurrence network and use its embedding for further generalizing author attributes.Standard graph feature engineering and network embedding methods were combined for constructing co-author recommender system formulated as link prediction problem and prediction of future graph structure.We evaluate our survey on the dataset containing temporal information on National Research University Higher School of Economics over twenty five years of research articles indexed in Russian Science Citation Index and Scopus.Our model of network representation shows better performance for stated binary classification tasks on several co-authorship networks.
BibTeX
@inproceedings{makarov2019peer,
  author = {Ilya Makarov and Olga Gerasimova and Pavel Sulimov and Leonid Zhukov and S Abu-El-Haija and B Perozzi and R Al-Rfou and S Adafre and M De Rijke and L Backstrom and J Leskovec and M Belkin and P Niyogi and H Cai and V Zheng and K Chang and S Cao and W Lu and Q Xu and N Cetorelli and S Peristiani and S Chang and W Han and J Tang and G.-J Qi and C Aggarwal and T Huang and H Chen and X Li and Z Huang and H Cho and Y Yu and P Cui and X Wang and J Pei and W Zhu and F Gao and K Musial and C Cooper and S Tsoka and S Gao and L Denoyer and P Gallinari and P Goyal and E Ferrara and A Grover and J Leskovec and Q He and J Pei and D Kifer and P Mitra and L Giles and X Huang and J Li and X Hu and X Kong and M Mao and W Wang and J Liu and B Xu and G Kossinets and D Watts and Q Le and T Mikolov and X Li and H Chen and Y Liang and Q Li and T Qian and D Liben-Nowell and J Kleinberg and Y Liu and Z Kou and L L and T Zhou and I Makarov and O Bulanov and O Gerasimova and N Meshcheryakova and I Karpov and L Zhukov and I Makarov and O Bulanov and L Zhukov and I Makarov and O Gerasimova and P Sulimov and K Korovina and L Zhukov and I Makarov and O Gerasimova and P Sulimov and L Zhukov and I Makarov and O Gerasimova and P Sulimov and L Zhukov and M Mcpherson and L Smith-Lovin and J Cook},
  title = {{Peer Review \#1 of "Dual network embedding for representing research interests in the link prediction problem on co-authorship networks (v0.2)"}},
  booktitle = {Openalex Unknown},
  year = {2019},
  doi = {10.7287/peerj-cs.172v0.2/reviews/1},
}
Figure