تقرير
Integrating Transductive And Inductive Embeddings Improves Link Prediction Accuracy
| العنوان: | Integrating Transductive And Inductive Embeddings Improves Link Prediction Accuracy |
|---|---|
| المؤلفون: | Gupta, Chitrank, Jain, Yash, De, Abir, Chakrabarti, Soumen |
| سنة النشر: | 2021 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Information Retrieval |
| الوصف: | In recent years, inductive graph embedding models, \emph{viz.}, graph neural networks (GNNs) have become increasingly accurate at link prediction (LP) in online social networks. The performance of such networks depends strongly on the input node features, which vary across networks and applications. Selecting appropriate node features remains application-dependent and generally an open question. Moreover, owing to privacy and ethical issues, use of personalized node features is often restricted. In fact, many publicly available data from online social network do not contain any node features (e.g., demography). In this work, we provide a comprehensive experimental analysis which shows that harnessing a transductive technique (e.g., Node2Vec) for obtaining initial node representations, after which an inductive node embedding technique takes over, leads to substantial improvements in link prediction accuracy. We demonstrate that, for a wide variety of GNN variants, node representation vectors obtained from Node2Vec serve as high quality input features to GNNs, thereby improving LP performance. Comment: 5 Pages, Accepted by CIKM 2021 |
| نوع الوثيقة: | Working Paper |
| DOI: | 10.1145/3459637.3482125 |
| URL الوصول: | http://arxiv.org/abs/2108.10108 |
| رقم الأكسشن: | edsarx.2108.10108 |
| قاعدة البيانات: | arXiv |
| DOI: | 10.1145/3459637.3482125 |
|---|