Learning Shared Vertex Representation in Heterogeneous Graphs with Convolutional Networks for Recommendation

Collaborative Filtering (CF) is among the most successful techniques in recommendation tasks. Recent works have shown a boost of performance of CF when introducing the pairwise relationships between users and items or among items (users) using interaction data. However, these works usually only utilize one kind of information, i.e., user preference in a user-item interaction matrix or item dependency in interaction sequences which can limit the recommendation performance. In this paper, we propose to mine three kinds of information (user preference, item dependency, and user similarity on behaviors) by converting interaction sequence data into multiple graphs (i.e., a user-item graph, an item-item graph, and a user-subseq graph). We design a novel graph convolutional network (PGCN) to learn shared representations of users and items with the three heterogeneous graphs. In our approach, a neighbor pooling and a convolution operation are designed to aggregate features of neighbors. Extensive experiments on two real-world datasets demonstrate that our graph convolution approaches outperform various competitive methods in terms of two metrics, and the heterogeneous graphs are proved effective for improving recommendation performance.

Download Full-text

Knowledge-aware Multi-modal Adaptive Graph Convolutional Networks for Fake News Detection

ACM Transactions on Multimedia Computing Communications and Applications ◽

10.1145/3451215 ◽

2021 ◽

Vol 17 (3) ◽

pp. 1-23

Author(s):

Shengsheng Qian ◽

Jun Hu ◽

Quan Fang ◽

Changsheng Xu

Keyword(s):

Social Media ◽

Visual Information ◽

Representation Learning ◽

Fake News ◽

Unified Framework ◽

Model Learning ◽

Convolutional Network ◽

Textual Information ◽

Convolutional Networks ◽

Real World Datasets

In this article, we focus on fake news detection task and aim to automatically identify the fake news from vast amount of social media posts. To date, many approaches have been proposed to detect fake news, which includes traditional learning methods and deep learning-based models. However, there are three existing challenges: (i) How to represent social media posts effectively, since the post content is various and highly complicated; (ii) how to propose a data-driven method to increase the flexibility of the model to deal with the samples in different contexts and news backgrounds; and (iii) how to fully utilize the additional auxiliary information (the background knowledge and multi-modal information) of posts for better representation learning. To tackle the above challenges, we propose a novel Knowledge-aware Multi-modal Adaptive Graph Convolutional Networks (KMAGCN) to capture the semantic representations by jointly modeling the textual information, knowledge concepts, and visual information into a unified framework for fake news detection. We model posts as graphs and use a knowledge-aware multi-modal adaptive graph learning principal for the effective feature learning. Compared with existing methods, the proposed KMAGCN addresses challenges from three aspects: (1) It models posts as graphs to capture the non-consecutive and long-range semantic relations; (2) it proposes a novel adaptive graph convolutional network to handle the variability of graph data; and (3) it leverages textual information, knowledge concepts and visual information jointly for model learning. We have conducted extensive experiments on three public real-world datasets and superior results demonstrate the effectiveness of KMAGCN compared with other state-of-the-art algorithms.

Download Full-text

Structure-Based Protein Function Prediction using Graph Convolutional Networks

10.1101/786236 ◽

2019 ◽

Cited By ~ 4

Author(s):

Vladimir Gligorijevic ◽

P. Douglas Renfrew ◽

Tomasz Kosciolek ◽

Julia Koehler Leman ◽

Daniel Berenberg ◽

...

Keyword(s):

Protein Function ◽

Sequence Data ◽

Protein Function Prediction ◽

General Structure ◽

Protein Structures ◽

Data Bank ◽

Function Prediction ◽

Convolutional Network ◽

Convolutional Networks ◽

Protein Functions

The large number of available sequences and the diversity of protein functions challenge current experimental and computational approaches to determining and predicting protein function. We present a deep learning Graph Convolutional Network (GCN) for predicting protein functions and concurrently identifying functionally important residues. This model is initially trained using experimentally determined structures from the Protein Data Bank (PDB) but has significant de-noising capability, with only a minor drop in performance observed when structure predictions are used. We take advantage of this denoising property to train the model on > 200,000 protein structures, including many homology-predicted structures, greatly expanding the reach and applications of the method. Our model learns general structure-function relationships by robustly predicting functions of proteins with ≤ 40% sequence identity to the training set. We show that our GCN architecture predicts functions more accurately than Convolutional Neural Networks trained on sequence data alone and previous competing methods. Using class activation mapping, we automatically identify structural regions at the residue-level that lead to each function prediction for every confidently predicted protein, advancing site-specific function prediction. We use our method to annotate PDB and SWISS-MODEL proteins, making several new confident function predictions spanning both fold and function classifications.

Download Full-text

Deep Residual Autoencoder with Multiscaling for Semantic Segmentation of Land-Use Images

Remote Sensing ◽

10.3390/rs11182142 ◽

2019 ◽

Vol 11 (18) ◽

pp. 2142 ◽

Cited By ~ 5

Author(s):

Lianfa Li

Keyword(s):

Land Use ◽

Deep Learning ◽

Semantic Segmentation ◽

Remotely Sensed ◽

Convolutional Network ◽

Convolutional Networks ◽

Residual Learning ◽

Fully Convolutional Networks ◽

Remotely Sensed Images ◽

Real World Datasets

Semantic segmentation is a fundamental means of extracting information from remotely sensed images at the pixel level. Deep learning has enabled considerable improvements in efficiency and accuracy of semantic segmentation of general images. Typical models range from benchmarks such as fully convolutional networks, U-Net, Micro-Net, and dilated residual networks to the more recently developed DeepLab 3+. However, many of these models were originally developed for segmentation of general or medical images and videos, and are not directly relevant to remotely sensed images. The studies of deep learning for semantic segmentation of remotely sensed images are limited. This paper presents a novel flexible autoencoder-based architecture of deep learning that makes extensive use of residual learning and multiscaling for robust semantic segmentation of remotely sensed land-use images. In this architecture, a deep residual autoencoder is generalized to a fully convolutional network in which residual connections are implemented within and between all encoding and decoding layers. Compared with the concatenated shortcuts in U-Net, these residual connections reduce the number of trainable parameters and improve the learning efficiency by enabling extensive backpropagation of errors. In addition, resizing or atrous spatial pyramid pooling (ASPP) can be leveraged to capture multiscale information from the input images to enhance the robustness to scale variations. The residual learning and multiscaling strategies improve the trained model’s generalizability, as demonstrated in the semantic segmentation of land-use types in two real-world datasets of remotely sensed images. Compared with U-Net, the proposed method improves the Jaccard index (JI) or the mean intersection over union (MIoU) by 4-11% in the training phase and by 3-9% in the validation and testing phases. With its flexible deep learning architecture, the proposed approach can be easily applied for and transferred to semantic segmentation of land-use variables and other surface variables of remotely sensed images.

Download Full-text

A directed link prediction method using graph convolutional network based on social ranking theory

Intelligent Data Analysis ◽

10.3233/ida-195006 ◽

2021 ◽

Vol 25 (3) ◽

pp. 739-757

Author(s):

Zheng Wu ◽

Hongchang Chen ◽

Jianpeng Zhang ◽

Shuxin Liu ◽

Ruiyang Huang ◽

...

Keyword(s):

Link Prediction ◽

Real Life ◽

Prediction Method ◽

Prediction Problem ◽

Social Ranking ◽

Convolutional Network ◽

Directional Information ◽

Convolutional Networks ◽

Ranking Theory ◽

Real World Datasets

Graph convolutional networks (GCN) have recently emerged as powerful node embedding methods in network analysis tasks. Particularly, GCNs have been successfully leveraged to tackle the challenging link prediction problem, aiming at predicting missing links that exist yet were not found. However, most of these models are oriented to undirected graphs, which are limited to certain real-life applications. Therefore, based on the social ranking theory, we extend the GCN to address the directed link prediction problem. Firstly, motivated by the reciprocated and unreciprocated nature of social ties, we separate nodes in the neighbor subgraph of the missing link into the same, a higher-ranked and a lower-ranked set. Then, based on the three kinds of node sets, we propose a method to correctly aggregate and propagate the directional information across layers of a GCN model. Empirical study on 8 real-world datasets shows that our proposed method is capable of reserving rich information related to directed link direction and consistently performs well on graphs from numerous domains.

Download Full-text

Metric Factorization with Item Cooccurrence for Recommendation

Symmetry ◽

10.3390/sym12040512 ◽

2020 ◽

Vol 12 (4) ◽

pp. 512 ◽

Cited By ~ 1

Author(s):

Honglin Dai ◽

Liejun Wang ◽

Jiwei Qin

Keyword(s):

Matrix Factorization ◽

Euclidean Distance ◽

Metric Learning ◽

Sparse Data ◽

Inner Product ◽

Interaction Matrix ◽

Latent Factors ◽

Rating Prediction ◽

Real World Datasets ◽

Item Item

In modern recommender systems, matrix factorization has been widely used to decompose the user–item matrix into user and item latent factors. However, the inner product in matrix factorization does not satisfy the triangle inequality, and the problem of sparse data is also encountered. In this paper, we propose a novel recommendation model, namely, metric factorization with item cooccurrence for recommendation (MFIC), which uses the Euclidean distance to jointly decompose the user–item interaction matrix and the item–item cooccurrence with shared latent factors. The item cooccurrence matrix is obtained from the colike matrix through the calculation of pointwise mutual information. The main contributions of this paper are as follows: (1) The MFIC model is not only suitable for rating prediction and item ranking, but can also well overcome the problem of sparse data. (2) This model incorporates the item cooccurrence matrix into metric learning so it can better learn the spatial positions of users and items. (3) Extensive experiments on a number of real-world datasets show that the proposed method substantially outperforms the compared algorithm in both rating prediction and item ranking.

Download Full-text

Graph Convolutional Network: Traffic Speed Prediction Fused with Traffic Flow Data

Sensors ◽

10.3390/s21196402 ◽

2021 ◽

Vol 21 (19) ◽

pp. 6402

Author(s):

Duanyang Liu ◽

Xinbo Xu ◽

Wei Xu ◽

Bingqian Zhu

Keyword(s):

Traffic Flow ◽

Intelligent Transportation Systems ◽

Transportation Systems ◽

Flow Data ◽

Convolutional Network ◽

Convolutional Networks ◽

Speed Prediction ◽

Traffic Speed ◽

Real World Datasets ◽

Relationship Of

Traffic speed prediction plays an important role in intelligent transportation systems, and many approaches have been proposed over recent decades. In recent years, methods using graph convolutional networks (GCNs) have been more promising, which can extract the spatiality of traffic networks and achieve a better prediction performance than others. However, these methods only use inaccurate historical data of traffic speed to forecast, which decreases the prediction accuracy to a certain degree. Moreover, they ignore the influence of dynamic traffic on spatial relationships and merely consider the static spatial dependency. In this paper, we present a novel graph convolutional network model called FSTGCN to solve these problems, where the model adopts the full convolutional structure and avoids repeated iterations. Specifically, because traffic flow has a mapping relationship with traffic speed and its values are more exact, we fused historical traffic flow data into the forecasting model in order to reduce the prediction error. Meanwhile, we analyzed the covariance relationship of the traffic flow between road segments and designed the dynamic adjacency matrix, which can capture the dynamic spatial correlation of the traffic network. Lastly, we conducted experiments on two real-world datasets and prove that our model can outperform state-of-the-art traffic speed prediction.

Download Full-text

GCNEXT: graph convolutional network with expanded balance theory for fraudulent user detection

Social Network Analysis and Mining ◽

10.1007/s13278-020-00697-w ◽

2020 ◽

Vol 10 (1) ◽

Author(s):

Wataru Kudo ◽

Mao Nishiguchi ◽

Fujio Toriumi

Keyword(s):

Real World ◽

Experimental Results ◽

Balance Theory ◽

Convolutional Network ◽

Convolutional Networks ◽

End To End ◽

Real World Datasets ◽

Credible Information ◽

Insight Into

AbstractRating platforms provide users with useful information on products or other users. However, fake ratings are sometimes generated by fraudulent users. In this paper, we tackle the task of fraudulent user detection on rating platforms. We propose GCNEXT (Graph Convolutional Network with Expended Balance Theory), an end-to-end framework based on graph convolutional networks (GCNs) and expanded balance theory, which properly incorporates both the signs and directions of edges. The experimental results on seven real-world datasets show that the proposed framework performs better, or even best, in most settings. In particular, this framework shows remarkable stability in inductive settings, which is associated with the detection of new fraudulent users on rating platforms. Furthermore, using expanded balance theory, we provide new insight into the behavior of users in rating networks that fraudulent users form a faction to deal with the negative ratings from other users. The owner of a rating platform can detect fraudulent users earlier and constantly provide users with more credible information by using the proposed framework.

Download Full-text

Efficient End-to-End Sentence-Level Lipreading with Temporal Convolutional Networks

Applied Sciences ◽

10.3390/app11156975 ◽

2021 ◽

Vol 11 (15) ◽

pp. 6975

Author(s):

Tao Zhang ◽

Lun He ◽

Xudong Li ◽

Guoqing Feng

Keyword(s):

Performance Improvement ◽

State Of The Art ◽

Error Rates ◽

Convolutional Network ◽

Convolutional Networks ◽

Sentence Level ◽

End To End ◽

High Level ◽

Improved Accuracy ◽

Talking Face

Lipreading aims to recognize sentences being spoken by a talking face. In recent years, the lipreading method has achieved a high level of accuracy on large datasets and made breakthrough progress. However, lipreading is still far from being solved, and existing methods tend to have high error rates on the wild data and have the defects of disappearing training gradient and slow convergence. To overcome these problems, we proposed an efficient end-to-end sentence-level lipreading model, using an encoder based on a 3D convolutional network, ResNet50, Temporal Convolutional Network (TCN), and a CTC objective function as the decoder. More importantly, the proposed architecture incorporates TCN as a feature learner to decode feature. It can partly eliminate the defects of RNN (LSTM, GRU) gradient disappearance and insufficient performance, and this yields notable performance improvement as well as faster convergence. Experiments show that the training and convergence speed are 50% faster than the state-of-the-art method, and improved accuracy by 2.4% on the GRID dataset.

Download Full-text

High-Fidelity Simulated Players for Interactive Narrative Planning

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence ◽

10.24963/ijcai.2018/540 ◽

2018 ◽

Author(s):

Pengcheng Wang ◽

Jonathan Rowe ◽

Wookhee Min ◽

Bradford Mott ◽

James Lester

Keyword(s):

State Of The Art ◽

Data Driven ◽

High Fidelity ◽

Interactive Narrative ◽

Interaction Data ◽

Convolutional Networks ◽

Novel Approach ◽

Adaptation Policies ◽

Narrative Planning ◽

Prior State

Interactive narrative planning offers significant potential for creating adaptive gameplay experiences. While data-driven techniques have been devised that utilize player interaction data to induce policies for interactive narrative planners, they require enormously large gameplay datasets. A promising approach to addressing this challenge is creating simulated players whose behaviors closely approximate those of human players. In this paper, we propose a novel approach to generating high-fidelity simulated players based on deep recurrent highway networks and deep convolutional networks. Empirical results demonstrate that the proposed models significantly outperform the prior state-of-the-art in generating high-fidelity simulated player models that accurately imitate human players’ narrative interactions. Using the high-fidelity simulated player models, we show the advantage of more exploratory reinforcement learning methods for deriving generalizable narrative adaptation policies.

Download Full-text

Evolutionary Social Poisson Factorizationfor Temporal Recommendation

International Journal of Computational Intelligence Systems ◽

10.1007/s44196-021-00022-z ◽

2021 ◽

Vol 14 (1) ◽

Author(s):

ChunYan Yin ◽

YongHeng Chen ◽

Wanli Zuo

Keyword(s):

Recommendation Systems ◽

Predictive Performance ◽

User Preferences ◽

Social Model ◽

User Preference ◽

Latent Factors ◽

Factorization Model ◽

Real World Datasets ◽

Item Attributes ◽

The Impact

AbstractPreference-based recommendation systems analyze user-item interactions to reveal latent factors that explain our latent preferences for items and form personalized recommendations based on the behavior of others with similar tastes. Most of the works in the recommendation systems literature have been developed under the assumption that user preference is a static pattern, although user preferences and item attributes may be changed through time. To achieve this goal, we develop an Evolutionary Social Poisson Factorization (EPF$$\_$$ _ Social) model, a new Bayesian factorization model that can effectively model the smoothly drifting latent factors using Conjugate Gamma–Markov chains. Otherwise, EPF$$\_$$ _ Social can obtain the impact of friends on social network for user’ latent preferences. We studied our models with two large real-world datasets, and demonstrated that our model gives better predictive performance than state-of-the-art static factorization models.

Download Full-text