Link Prediction and Node Classification Based on Multitask Graph Autoencoder

Abstract

The goal of network representation learning is to extract deep-level abstraction from data features that can also be viewed as a process of transforming the high-dimensional data to low-dimensional features. Learning the mapping functions between two vector spaces is an essential problem. In this paper, we propose a new similarity index based on traditional machine learning, which integrates the concepts of common neighbor, local path, and preferential attachment. Furthermore, for applying the link prediction methods to the field of node classification, we have innovatively established an architecture named multitask graph autoencoder. Specifically, in the context of structural deep network embedding, the architecture designs a framework of high-order loss function by calculating the node similarity from multiple angles so that the model can make up for the deficiency of the second-order loss function. Through the parameter fine-tuning, the high-order loss function is introduced into the optimized autoencoder. Proved by the effective experiments, the framework is generally applicable to the majority of classical similarity indexes.

Classification of Artistic Styles of Chinese Art Paintings Based on the CNN Model

“…e DEFC embedded mapping framework [9][10][11][12][13][14][15] follows the principle of "friendship is close, hostility is far," and it introduces two parameters, C F and E D , to control the embedding process; then,…”

“…e DEFC (data embedding framework for classi cation) [9][10][11][12][13][14][15] takes the relevant features obtained through the selective conversion of data samples as the input of the algorithm. Although it is much improved than the existing algorithm, it does not consider the in uence of feature importance on the calculation of similar features between objects in the process of calculating similar features between objects, resulting in the restriction of classi cation results.…”

“…e experimental results show that although the mapping matrices obtained by solving formula (11) or (12) are different, the classification results of the datasets obtained after the mapping calculation are the same.…”

Link-Backdoor: Backdoor Attack on Link Prediction via Node Injection

“…GNNs-based methods. In terms of the nonlinear and hierarchical nature of neural networks, the GNNs-based methods [14] [32], [29], [26], [5] adopt graph neural networks to capture the potential features vectors of nodes. Based on the local enclosing subgraphs, Zhang and Chen [39] absorbed multiple types of information from subgraph structures and latent node features to learn general graph structure features.…”

“…Besides, embeddingbased methods [27], [10], [31] direct their efforts at adopting random walking to characterize the nodes. With the rapid development and wide applications of deep learning methods [16], [18], the graph neural networks (GNNs) methods have shown promising performance in link prediction [15], [29], [32], [5]. They take advantage of the nonlinear and hierarchical nature of neural networks to capture the potential characteristic vectors of nodes, which makes prediction accuracy better than non-GNNs based methods.…”