A Study of Dynamic Heterogeneous Network Prediction based on DyHATR-Skip Embedding Fusion

DOI

10.2991/978-94-6463-108-1_83How to use a DOI?

Keywords

Dynamic heterogeneous networks; DyHATR model; Skip-gram model; Embedding fusion

Abstract

To solve the problem that the single dynamic heterogeneous network embedding method (DyHATR) cannot capture the node features accurately and adequately, which leads to the low efficiency of the final link prediction. This paper proposes to solve this problem by using the DyHATR based on the Skipgram method (DyHATR-Skip): (1) Generating word embedding by using the Skip-gram model in Word2vec; (2) Fusing the generated word embedding with the node embedding generated by DyHATR for splicing fusion, which is named as DyHATR-Skip. The method generates new node embedding by DyHATR and Skip-gram models. The experimental results show that the DyHATR-Skip method proposed in this paper performs better than the single DyHATR method. In the DyHATR-Skip method, AUROC improves 0.07, 0.01, 0.05 and AUPRC improves 0.07, 0.01, 0.03 on Twitter, Math-Overflow and EComm datasets respectively. Therefore, the DyHATR-Skip method proposed in this paper can capture node features and generate node embedding more fully and accurately compared to single network embedding methods and has better performance in dynamic link prediction. But since words and vectors are one-to-one in Word2vec, DyHATR-Skip has some limitations for multisense words and complex datasets.

Copyright

© 2022 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhaoke Li
AU  - Shuwei Xu
AU  - Gaofei Si
AU  - Jingyun Zhang
PY  - 2022
DA  - 2022/12/30
TI  - A Study of Dynamic Heterogeneous Network Prediction based on DyHATR-Skip Embedding Fusion
BT  - Proceedings of the 2022 International Conference on Computer Science, Information Engineering and Digital Economy (CSIEDE 2022)
PB  - Atlantis Press
SP  - 749
EP  - 754
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-108-1_83
DO  - 10.2991/978-94-6463-108-1_83
ID  - Li2022
ER  -