Airport delay prediction model based on regional residual and LSTM network by Jingyi QU, Meng YE, Xing QU

“…Nowadays,the civil aviation industry has a high precision requirement of airport delay prediction,so an airport delay prediction model based on the RR-LSTM network was proposed.Firstly,the airport information,meteorological information and related flight information were integrated.Then,the RR-LSTM network was used to extract the features of the fused airport data set.Finally,the Softmax classifier was adopted to classify and predict the airport delay.The proposed RR-LSTM network model can not only extract the time correlation of airport delay data effectively,but also avoid the gradient disappearance problem of deep LSTM network.The experimental results indicate that the RR-LSTM network model has a prediction accuracy of 95.52%,which achieves better prediction results than the traditional network model.The prediction accuracy can be improved about 11% by fusing the weather information and the flight information of the airport.…”
Get full text

Improved Convolutional Neural Image Recognition Algorithm based on LeNet-5 by Lijie Zhou, Weihai Yu

“…Based on convolution operation, pooling operation, softmax classifier, and network optimization algorithm in improved convolutional neural network of LeNet-5, this paper conducts image recognition experiments on handwritten digits and face datasets, respectively. …”
Get full text

Intelligent System of Somatosensory Music Therapy Information Feedback in Deep Learning Environment by Nan Zhao

“…Based on the theoretical basis of Deep Belief Network (DBN) in deep learning, this paper proposes a method that combines the optimized Restricted Boltzmann machine (RBM) feature extraction model with the softmax classification algorithm. Brain wave tracking analysis is performed on children with autism who have received different music perception treatments to improve classification accuracy and achieve the purpose of accurately judging the condition. …”
Get full text

Quality of service optimization algorithm based on deep reinforcement learning in software defined network by Cenhuishan LIAO, Junyan CHEN, Guanping LIANG, Xiaolan XIE, Xiaoye LU

“…Deep reinforcement learning has strong abilities of decision-making and generalization and often applies to the quality of service (QoS) optimization in software defined network (SDN).However, traditional deep reinforcement learning algorithms have problems such as slow convergence and instability.An algorithm of quality of service optimization algorithm of based on deep reinforcement learning (AQSDRL) was proposed to solve the QoS problem of SDN in the data center network (DCN) applications.AQSDRL introduces the softmax deep double deterministic policy gradient (SD3) algorithm for model training, and a SumTree-based prioritized empirical replay mechanism was used to optimize the SD3 algorithm.The samples with more significant temporal-difference error (TD-error) were extracted with higher probability to train the neural network, effectively improving the convergence speed and stability of the algorithm.The experimental results show that the proposed AQSDRL effectively reduces the network transmission delay and improves the load balancing performance of the network than the existing deep reinforcement learning algorithms.…”
Get full text

Hierarchical Deep Learning for Bearing Fault Detection in BLDC Motors Using Time-Frequency Analysis by Ahmed K. Ali, Wathiq Rafa Abed

“…We detect the targets using three patterns generated by the convolutional layers, utilizing three deep network structures, each containing a softmax classifier that identifies one pattern out of three. …”
Get full text

FAULT DIAGNOSIS OF ROLLING BEARING BASED ON UNSUPERVISED FEATURE ALIGNMENT by ZHANG Tao, JIA Qian, XIN YueJie

“…By minimizing the second-order statistical difference and the maximum mean difference of the joint distribution, the differences in the feature distributions of the source domain and target domain were reduced, and the common features of the two domains were extracted. Finally, add Softmax classification layer to realize fault status recognition of target data. …”
Get full text

Automated crowdturfing attack in Chinese user reviews by Li’na WANG, Xiaodong GUO, Run WANG

“…The text-oriented automated crowdturfing attack has a series of features such as low attack cost and strong concealment.This kind of attack can automatically generate a large number of fake reviews,with harmful effect on the healthy development of the user review community.In recent years,researchers have found that text-oriented crowdturfing attacks for the English review community,but there was few research work on automated crowdsourcing attacks in the Chinese review community.A Chinese character embedding LSTM model was proposed to automatically generate Chinese reviews with the aim of antomated crowdturfing attacks,which model trained by a combination with Chinese character embedding network,LSTM network and softmax dense network,and a temperature parameter T was designed to construct the attack model.In the experiment,more than 50 000 real user reviews were crawled from Taobao's online review platform to verify the effectiveness of the attack method.Experimental results show that the generated fake reviews can effectively fool linguistics-based classification detection approach and texts plagiarism detection approach.Besides,the massive manually evaluation experiments also demonstrate that the generated reviews with the proposed attack approach perform well in reality and diversity.…”
Get full text

Method based on contrastive incremental learning for fine-grained malicious traffic classification by Yifeng WANG, Yuanbo GUO, Qingli CHEN, Chen FANG, Renhao LIN, Yongliang ZHOU, Jiali MA

“…In order to protect against continuously emerging unknown threats, a new method based on contrastive incremental learning for fine-grained malicious traffic classification was proposed.The proposed method was based on variational auto-encoder (VAE) and extreme value theory (EVT), and the high accuracy could be achieved in known, few-shot and unknown malicious classes and new classes were also identified without using a large number of old task samples, which met the demand of storage and time cost in incremental learning scenarios.Specifically, the contrastive learning was integrated into the encoder of VAE, and the A-Softmax was used for known and few-shot malicious traffic classification, EVT and the decoder of VAE were used for unknown malicious traffic recognition, all classes could be recognized without a lot of old samples when learning new tasks by using VAE reconstruction and knowledge distillation methods.Experimental results indicate that the proposed method is efficient in known, few-shot and unknown malicious classes, and has greatly reduced the forgetting speed of old knowledge in incremental learning scenarios.…”
Get full text

Fault Diagnosis Analysis of Variable Working Condition Gearbox based on DWAE and GRUNN Combination Model by Chuanhui Liu, Xiaojing Chen, Xiaoxiao Hou, Qiang Zhu, Yong Dong

“…The Adam and Dropout methods are used for training， and the softmax classifier is used to accurately identify the operating state of the gearbox under different working conditions. …”
Get full text

Evidence classification method of chat text based on DSR and BGRU model by Yu ZHANG, Binglong LI, Xuejuan LI, Heyu ZHANG

“…It is always unlikely to efficiently identify and extract chat text evidence related to criminal events, due to the complex semantics such as “slang” in the chat content and the huge amount of chat text data generated by social software such as instant messaging.Based on this motivation, a chat text evidence classification model (DSR-BGRU) based on the DSR (dynamic semantic representation) model and the BGRU (bidirectional gated recurrent unit) model was proposed.The chat text data was pre-processed to preserve the characteristics of the criminal field.Then a multi-layer chat text feature extraction and classification model using the Keras framework was proposed.With the text matrix composed of vector representation of words in the DSR model as the input vector, the input layer of the DSR model featured the chat text from the semantic level.Then the hidden layer of the BGRU model extracted the context characteristics of the text composed of the word vectors.The softmax classification layer recognized and extracted the chat text evidence.The experimental results show that the proposed DSR-BGRU can more accurately identify and extract chat records compared with other models and methods for text classification, and it can also effectively extract the criminal text information from the chat information with the accuracy rate 92.06% and the F1 score 91.00%.…”
Get full text

Deep Learning for Person Reidentification Using Support Vector Machines by Mengyu Xu, Zhenmin Tang, Yazhou Yao, Lingxiang Yao, Huafeng Liu, Jingsong Xu

“…Different from previous works, we represent the pairs of pedestrian images as new resized input and use linear Support Vector Machine to replace softmax activation function for similarity learning. …”
Get full text

Graph Convolutional Network for Word Sense Disambiguation by Chun-Xiang Zhang, Rui Liu, Xue-Yao Gao, Bo Yu

“…GCN is used to fuse features of a node and its neighbors, and the softmax function is applied to determine the semantic category of the ambiguous word. …”
Get full text

A multi-source threat intelligence confidence value evaluation method based on machine learning by Hansheng LIU, Hongyu TANG, Mingxia BO, Jianfeng NIU, Tianbo LI, Lingxiao LI

“…During the collection process of multi-source threat intelligence,it is very hard for the intelligence center to make a scientific decision to massive intelligence because the data value density is low,the intelligence repeatabil-ity is high,and the ineffective time is very short,etc.Based on those problems,a new multi-source threat intelligence confidence value evaluation method was put forward based on machine learning.First of all,according to the STIX intelligence standard format,a multi-source intelligence data standardization process was designed.Secondly,ac-cording to the characteristic of data,14 characteristics were extracted from four dimensions of publishing time,source,intelligence content and blacklist matching degree to be the basis of determining the intelligence reliability.After getting the feature encoding,an intelligence confidence value evaluation model was designed based on deep neural network algorithm and Softmax classifier.Backward propagation algorithm was also used to minimize recon-struction error.Last but not least,according to the 2 000 open source marked sample data,k-ford cross-validation method was used to evaluate the model and get an average of 91.37% macro-P rate and 84.89% macro-R rate.It was a good reference for multi-source threat intelligence confidence evaluation.…”
Get full text

Deep Learning-Based Dzongkha Handwritten Digit Classification by Yonten Jamtsho, Pema Yangden, Sonam Wangmo, Nima Dema

“…In the study, the 3 layer set of CONV → ReLU → POOL, followed by a fully connected layer, dropout layer, and softmax function were used to train the digit. In the dataset, each class (0-9) contains 1500 images which are split into train, validation, and test sets: 70:20:10. …”
Get full text

Evaluation of physical fitness parameters of hearing impaired adolescents who are active and non-active in sports by A. Uludag, Erkan Calıskan, Ozge Caglar, Tulin Sepetcı

Get full text

Bearing Small Sample Fault Diagnosis based on Data Generation and Transfer Learning by Dinghui Wu, Qin Fang, Chuyi Wu

“…First， the source dataset is trained on the training network to obtain the source model， and then a small amount of drive-end data is used as the target dataset to fine-tuning to obtain the target model. Finally， the Softmax function is used for fault diagnosis on the output of the fully-connected layer of the target model. …”
Get full text

An Enhancement Deep Feature Extraction Method for Bearing Fault Diagnosis Based on Kernel Function and Autoencoder by Fengtao Wang, Bosen Dun, Xiaofei Liu, Yuhang Xue, Hongkun Li, Qingkai Han

“…Subsequently, a deep neural network is constructed with one KAE and multiple AEs to extract inherent features layer by layer. Finally, softmax is adopted as the classifier to accurately identify different bearing faults, and error backpropagation algorithm is used to fine-tune the model parameters. …”
Get full text

Unbalanced protocol recognition method based on improved residual U-Net by Jisheng WU, Zheng HONG, Tiantian MA

“…An unbalanced protocol recognition method based on the improved Residual U-Net was proposed to solve the challenge of network security posed by the increasing network attacks with the continuous development of the Internet.In the captured network traffic, a small proportion is constituted by malicious traffic, typically utilizing minority protocols.However, existing protocol recognition methods struggle to accurately identify these minority protocols when the class distribution of the protocol data is imbalanced.To address this issue, an unbalanced protocol recognition method was proposed, which utilized the improved Residual U-Net, incorporating a novel activation function and the Squeeze-and-Excitation Networks (SE-Net) to enhance the feature extraction capability.The loss function employed in the proposed model was the weighted Dice loss function.In cases where the recognition accuracies of the minority protocols were low, the loss function value would be high.Consequently, the optimization direction of the model would be dominated by the minority protocols, resulting in improved recognition accuracies for them.During the protocol recognition process, the network flow was extracted from the network traffic and preprocessed to convert it into a one-dimensional matrix.Subsequently, the protocol recognition model extracted the features of the protocol data, and the Softmax classifier predicted the protocol types.Experimental results demonstrate that the proposed protocol recognition model achieves more accurate recognition of the minority protocols compared to the comparison model, while also improving the recognition accuracies of the majority protocols.…”
Get full text

Mass Laplacian Discriminant Analysis and Its Application in Gear Fault Diagnosis by Guangbin Wang, Ying Lv, Tengqiang Wang, Xiaohui Wang, Huanke Cheng

“…Finally, based on the mapping function, the eigenvalues of the training data and the test data are calculated, and the softmax algorithm is used to classify the test data. …”
Get full text

Rolling Bearing Fault Diagnosis Based on STFT-Deep Learning and Sound Signals by Hongmei Liu, Lianfeng Li, Jian Ma

“…After spectrograms are obtained by short-time Fourier transform, stacked sparse autoencoder is employed to automatically extract the fault features, and softmax regression is adopted as the method for classifying the fault modes. …”
Get full text