Recommendation algorithm based on attributed multiplex heterogeneous network

In the field of deep learning, the processing of large network models on billions or even tens of billions of nodes and numerous edge types is still flawed, and the accuracy of recommendations is greatly compromised when large network embeddings are applied to recommendation systems. To solve the problem of inaccurate recommendations caused by processing deficiencies in large networks, this paper combines the attributed multiplex heterogeneous network with the attention mechanism that introduces the softsign and sigmoid function characteristics and derives a new framework SSN_GATNE-T (S represents the softsign function, SN represents the attention mechanism introduced by the Softsign function, and GATNE-T represents the transductive embeddings learning for attribute multiple heterogeneous networks). The attributed multiplex heterogeneous network can help obtain more user-item information with more attributes. No matter how many nodes and types are included in the model, our model can handle it well, and the improved attention mechanism can help annotations to obtain more useful information via a combination of the two. This can help to mine more potential information to improve the recommendation effect; in addition, the application of the softsign function in the fully connected layer of the model can better reduce the loss of potential user information, which can be used for accurate recommendation by the model. Using the Adam optimizer to optimize the model can not only make our model converge faster, but it is also very helpful for model tuning. The proposed framework SSN_GATNE-T was tested for two different types of datasets, Amazon and YouTube, using three evaluation indices, ROC-AUC (receiver operating characteristic-area under curve), PR-AUC (precision recall-area under curve) and F1 (F1-score), and found that SSN_GATNE-T improved on all three evaluation indices compared to the mainstream recommendation models currently in existence. This not only demonstrates that the framework can deal well with the shortcomings of obtaining accurate interaction information due to the presence of a large number of nodes and edge types of the embedding of large network models, but also demonstrates the effectiveness of addressing the shortcomings of large networks to improve recommendation performance. In addition, the model is also a good solution to the cold start problem.

Download Full-text

Higher-order simplicial synchronization of coupled topological signals

Communications Physics ◽

10.1038/s42005-021-00605-4 ◽

2021 ◽

Vol 4 (1) ◽

Author(s):

Reza Ghorbanchian ◽

Juan G. Restrepo ◽

Joaquín J. Torres ◽

Ginestra Bianconi

Keyword(s):

Algebraic Topology ◽

Network Models ◽

Simplicial Complexes ◽

Higher Order ◽

Large Network ◽

Underlying Network ◽

Discontinuous Phase ◽

Fully Connected ◽

The Brain ◽

Fully Connected Networks

AbstractSimplicial complexes capture the underlying network topology and geometry of complex systems ranging from the brain to social networks. Here we show that algebraic topology is a fundamental tool to capture the higher-order dynamics of simplicial complexes. In particular we consider topological signals, i.e., dynamical signals defined on simplices of different dimension, here taken to be nodes and links for simplicity. We show that coupling between signals defined on nodes and links leads to explosive topological synchronization in which phases defined on nodes synchronize simultaneously to phases defined on links at a discontinuous phase transition. We study the model on real connectomes and on simplicial complexes and network models. Finally, we provide a comprehensive theoretical approach that captures this transition on fully connected networks and on random networks treated within the annealed approximation, establishing the conditions for observing a closed hysteresis loop in the large network limit.

Download Full-text

Machine Learning Models for Diagnostic Classiﬁcation of Hepatitis C Tests

Frontiers in Health Informatics ◽

10.30699/fhi.v10i1.274 ◽

2021 ◽

Vol 10 (1) ◽

pp. 70

Author(s):

Oladosu Oyebisi Oladimeji ◽

Abimbola Oladimeji ◽

Oladimeji Olayanju

Keyword(s):

Hepatitis C ◽

Area Under Curve ◽

Operating Characteristic ◽

Matthews Correlation Coefficient ◽

Receiver Operating Characteristic Area ◽

Precision Recall Curve ◽

Characteristic Area ◽

F Measure ◽

Machine Learning Models ◽

Recall Curve

Introduction: Hepatitis C is a chronic infection caused by hepatitis c virus - a blood borne virus. Therefore, the infection occurs through exposure to small quantities of blood. It has been estimated by World Health Organization (WHO) to have affected 71 million people worldwide. This infection costs individual, groups and government a lot because no vaccine has been gotten yet for the treatment. This disease is likely to continue to affect more people because it’s long asymptotic phase which makes its early detection not feasible.Material and Methods: In this study, we have presented machine learning models to automatically classify the diagnosis test of hepatitis and also ranked the test features in order to know how they contribute to the classification which help in decision making process by the health care industry. The synthetic minority oversampling technique (SMOTE) was used to solve the problem of imbalance dataset.Results: The models were evaluated based on metrics such as Matthews correlation coefficient, F-measure, Precision-Recall curve and Receiver Operating Characteristic Area Under Curve. We found that using SMOTE techniques helped raise performance of the predictive models. Also, random forest (RF) had the best performance based on Matthews correlation coefficient (0.99), F-measure (0.99), Precision-Recall curve (1.00) and Receiver Operating Characteristic Area Under Curve (0.99).Conclusion: This discovery has the potential to impact on clinical practice, when health workers aim at classifying diagnosis result of disease at its early stage.

Download Full-text

An Empirical Study on the Countermeasures of Implementing 5G Multimedia Network Technology in College Education

Mobile Information Systems ◽

10.1155/2021/2547648 ◽

2021 ◽

Vol 2021 ◽

pp. 1-14

Author(s):

Ke Li ◽

Sang-Bing Tsai

Keyword(s):

Physical Education ◽

Heterogeneous Network ◽

Evaluation System ◽

Evaluation Process ◽

Representation Learning ◽

Attention Mechanism ◽

Teaching Model ◽

Education Evaluation ◽

Network Representation ◽

Multimedia Network

Aiming at the problem of 5G multimedia heterogeneous multimodal network representation learning, this paper proposes a collaborative multimodal heterogeneous network representation learning method based on attention mechanism. This method learns different representations for nodes based on heterogeneous network structure information and multimodal content and designs an attention mechanism to learn weights for different representations to fuse them to obtain robust node representations. Combining the general process of exploring the college physical education model and the characteristics of the multimedia network classroom environment, this article constructs the process of exploring the college physical education teaching model of the multimedia network classroom. Through the research and practice of the inquiry college physical education teaching model in the multimedia network classroom, it is verified that the implementation of the inquiry college physical education teaching in the multimedia network classroom can effectively influence and increase the students’ interest in learning and stimulate the students’ inner learning motivation. Through the guidance and training of teachers, a variety of disciplines can be used to carry out college physical education in multimedia network classrooms, so that the integration between courses can be truly realized, with the aim that all courses can share the excellent results brought by the development of modern education technology. More educators understand, accept, and participate in the practice of college physical education based on multimedia network classrooms and better serve the education of college physical education. The construction of the college physical education evaluation system should be combined with the characteristics of the 5G multimedia network era. The evaluation process includes data collection, data analysis, result output, and result feedback. Each link is an indispensable part of the college physical education evaluation process. Based on the relevant knowledge of the 5G multimedia network, the evaluation indicators determined in this study can basically reflect the various elements of the physical education process in colleges and universities. The distribution of index weight coefficients is more scientific and reasonable. Compared with the current system, the college physical education evaluation system constructed by exploration has a certain degree of objectivity and scientificity. Therefore, it is feasible to apply the 5G multimedia network to the evaluation of college physical education.

Download Full-text

A Curve Shaped Description of Large Networks, with an Application to the Evaluation of Network Models

PLoS ONE ◽

10.1371/journal.pone.0019784 ◽

2011 ◽

Vol 6 (5) ◽

pp. e19784 ◽

Cited By ~ 1

Author(s):

Xianchuang Su ◽

Xiaogang Jin ◽

Yong Min ◽

Linjian Mo ◽

Jiangang Yang

Keyword(s):

Network Models ◽

Large Networks

Download Full-text

Loss-Based Attention for Deep Multiple Instance Learning

Proceedings of the AAAI Conference on Artificial Intelligence ◽

10.1609/aaai.v34i04.6030 ◽

2020 ◽

Vol 34 (04) ◽

pp. 5742-5749

Author(s):

Xiaoshuang Shi ◽

Fuyong Xing ◽

Yuanpu Xie ◽

Zizhao Zhang ◽

Lei Cui ◽

...

Keyword(s):

State Of The Art ◽

Classification Performance ◽

Multiple Instance Learning ◽

Attention Mechanism ◽

Cross Entropy ◽

General Category ◽

Source Codes ◽

Model Generalization ◽

Fully Connected ◽

Entropy Functions

Although attention mechanisms have been widely used in deep learning for many tasks, they are rarely utilized to solve multiple instance learning (MIL) problems, where only a general category label is given for multiple instances contained in one bag. Additionally, previous deep MIL methods firstly utilize the attention mechanism to learn instance weights and then employ a fully connected layer to predict the bag label, so that the bag prediction is largely determined by the effectiveness of learned instance weights. To alleviate this issue, in this paper, we propose a novel loss based attention mechanism, which simultaneously learns instance weights and predictions, and bag predictions for deep multiple instance learning. Specifically, it calculates instance weights based on the loss function, e.g. softmax+cross-entropy, and shares the parameters with the fully connected layer, which is to predict instance and bag predictions. Additionally, a regularization term consisting of learned weights and cross-entropy functions is utilized to boost the recall of instances, and a consistency cost is used to smooth the training process of neural networks for boosting the model generalization performance. Extensive experiments on multiple types of benchmark databases demonstrate that the proposed attention mechanism is a general, effective and efficient framework, which can achieve superior bag and image classification performance over other state-of-the-art MIL methods, with obtaining higher instance precision and recall than previous attention mechanisms. Source codes are available on https://github.com/xsshi2015/Loss-Attention.

Download Full-text

Scaling of average weighted shortest path and average receiving time on weighted expanded Koch networks

International Journal of Modern Physics B ◽

10.1142/s0217979214501112 ◽

2014 ◽

Vol 28 (17) ◽

pp. 1450111 ◽

Cited By ~ 9

Author(s):

Zikai Wu ◽

Baoyu Hou ◽

Hongjuan Zhang ◽

Feng Jin

Keyword(s):

Shortest Path ◽

Linear Dependence ◽

Path Length ◽

Network Models ◽

Structural Features ◽

The Other ◽

Large Network ◽

Dynamical Processes ◽

Current Generation ◽

Evolutionary Step

Deterministic network models have been attractive media for discussing dynamical processes' dependence on network structural features. On the other hand, the heterogeneity of weights affect dynamical processes taking place on networks. In this paper, we present a family of weighted expanded Koch networks based on Koch networks. They originate from a r-polygon, and each node of current generation produces m r-polygons including the node and whose weighted edges are scaled by factor w in subsequent evolutionary step. We derive closed-form expressions for average weighted shortest path length (AWSP). In large network, AWSP stays bounded with network order growing (0 < w < 1). Then, we focus on a special random walks and trapping issue on the networks. In more detail, we calculate exactly the average receiving time (ART). ART exhibits a sub-linear dependence on network order (0 < w < 1), which implies that nontrivial weighted expanded Koch networks are more efficient than un-weighted expanded Koch networks in receiving information. Besides, efficiency of receiving information at hub nodes is also dependent on parameters m and r. These findings may pave the way for controlling information transportation on general weighted networks.

Download Full-text

Deep Neural Network Compression for Plant Disease Recognition

Symmetry ◽

10.3390/sym13101769 ◽

2021 ◽

Vol 13 (10) ◽

pp. 1769

Author(s):

Ruiqing Wang ◽

Wu Zhang ◽

Jiuyang Ding ◽

Meng Xia ◽

Mengjian Wang ◽

...

Keyword(s):

Plant Disease ◽

Resource Constraints ◽

Network Models ◽

Performance Comparison ◽

Plant Diseases ◽

Compression Method ◽

Before And After ◽

Knowledge Distillation ◽

Network Compression ◽

Fully Connected

Deep neural networks (DNNs) have become the de facto standard for image recognition tasks, and their applications with respect to plant diseases have also obtained remarkable results. However, the large number of parameters and high computational complexities of these network models make them difficult to deploy on farms in remote areas. In this paper, focusing on the problems of resource constraints and plant diseases, we propose a DNN-based compression method. In order to reduce computational burden, this method uses lightweight fully connected layers to accelerate reasoning, pruning to remove redundant parameters and reduce multiply–accumulate operations, knowledge distillation instead of retraining to restore the lost accuracy, and then quantization to compress the size of the model further. After compressing the mainstream VGGNet and AlexNet models, the compressed versions are applied to the Plant Village dataset of plant disease images, and a performance comparison of the models before and after compression is obtained to verify the proposed method. The results show that the model can be compressed to 0.04 Mb with an accuracy of 97.09%. This experiment also proves the effectiveness of knowledge distillation during the pruning process, and compressed models are more efficient than prevalent lightweight models.

Download Full-text

Understanding fully-connected and convolution allayers in unsupervised learning using face images

10.5753/wvc.2019.7621 ◽

2019 ◽

Author(s):

Lucas Fontes Buzuti ◽

Carlos Eduardo Thomaz

Keyword(s):

Neural Network ◽

Deep Learning ◽

Unsupervised Learning ◽

Facial Expressions ◽

Multivariate Statistics ◽

Network Models ◽

Neural Network Models ◽

Face Images ◽

Convolutional Autoencoder ◽

Fully Connected

The goal of this paper is to implement and compare two unsupervised models of deep learning: Autoencoder and Convolutional Autoencoder. These neural network models have been trained to learn regularities in well-framed face images with different facial expressions. The Autoencoder's basic topology is addressed here, composed of encoding and decoding multilayers. This paper approaches these automatic codings using multivariate statistics to visually understand the bottleneck differences between the fully-connected and convolutional layers and the corresponding importance of the dropout strategy when applied in a model.

Download Full-text

A Self-Attention Mechanism Neural Network for Detection and Diagnosis of COVID-19 from Chest X-Ray Images

10.21203/rs.3.rs-577494/v1 ◽

2021 ◽

Author(s):

Bo Cheng ◽

Wei Xiang ◽

Ruhui Xue ◽

Hang Yang ◽

Laili Zhu

Keyword(s):

Neural Network ◽

Attention Mechanism ◽

X Rays ◽

Diagnostic Features ◽

Data Set ◽

X Ray ◽

Difficulty Breathing ◽

Enhancement Method ◽

Detection And Diagnosis ◽

Characteristic Area

Abstract The new type of coronavirus is called COVID-19. The virus can cause respiratory diseases, accompanied by cough, fever, difficulty breathing, and in severe cases, it can also cause symptoms such as pneumonia. It began to spread at the end of 2019 and has now spread to all parts of the world. The limited test kits and increasing number of cases encourage us to propose a deep learning model that can help radiologists and clinicians use chest X-rays to detect COVID-19 cases and show the diagnostic features of pneumonia. In this study, our methods are: 1) Propose a data enhancement method to increase the diversity of the data set, thereby improving the generalization performance of the network. 2) Using the deep convolutional neural network model DPN-SE, an attention mechanism is added on the basis of the DPN network, which greatly improves the performance of the network. 3) Use the lime interpretable library to mark the X-ray, the characteristic area on the medical image that is helpful for the doctor to make a diagnosis. The model we proposed can obtain better results with the least amount of data preprocessing given limited data. In general, the proposed method and model can effectively become a very useful tool for clinical practitioners and radiologists.

Download Full-text

Aspect-Level Sentiment Analysis Based on Position Features Using Multilevel Interactive Bidirectional GRU and Attention Mechanism

Discrete Dynamics in Nature and Society ◽

10.1155/2020/5824873 ◽

2020 ◽

Vol 2020 ◽

pp. 1-13

Author(s):

Xiaodi Wang ◽

Xiaoliang Chen ◽

Mingwei Tang ◽

Tian Yang ◽

Zhen Wang

Keyword(s):

Neural Network ◽

Sentiment Analysis ◽

Network Models ◽

Attention Mechanism ◽

Sentiment Classification ◽

Classification Model ◽

Neural Network Models ◽

Position Information ◽

Gated Recurrent Unit ◽

Relative Position Information

The aim of aspect-level sentiment analysis is to identify the sentiment polarity of a given target term in sentences. Existing neural network models provide a useful account of how to judge the polarity. However, context relative position information for the target terms is adversely ignored under the limitation of training datasets. Considering position features between words into the models can improve the accuracy of sentiment classification. Hence, this study proposes an improved classification model by combining multilevel interactive bidirectional Gated Recurrent Unit (GRU), attention mechanisms, and position features (MI-biGRU). Firstly, the position features of words in a sentence are initialized to enrich word embedding. Secondly, the approach extracts the features of target terms and context by using a well-constructed multilevel interactive bidirectional neural network. Thirdly, an attention mechanism is introduced so that the model can pay greater attention to those words that are important for sentiment analysis. Finally, four classic sentiment classification datasets are used to deal with aspect-level tasks. Experimental results indicate that there is a correlation between the multilevel interactive attention network and the position features. MI-biGRU can obviously improve the performance of classification.

Download Full-text