Filters








26,133 Hits in 6.2 sec

Similarity Detection for Higher-Order Structure of DNA Sequences

Nguyen Thi Ngoc Anh, Ho Phan Hieu, Tran Anh Kiet, Vo Trung Hung
2017 Journal of Science and Technology Issue on Information and Communications Technology  
The contributions will be distributed along two main thrusts of effectiveness; including latent modeling setting for imputing missing values based on the High-Order Kalman Filter and feature extraction  ...  based on Tensor Discriminative Feature Extraction.  ...  Acknowledgment This research is funded by Funds for Science and Technology Development of the University of Danang under grant number B2017-DN01-07 and B2017-DN03-07.  ... 
doi:10.31130/jst.2017.51 fatcat:w7gl57cayzbvvgknnnqwdoh3ga

Intrusion Detection System Using Deep Learning and Its Application to Wi-Fi Network

Kwangjo KIM
2020 IEICE transactions on information and systems  
For this, the author has suggested the concept of the Deep-Feature Extraction and Selection (D-FES).  ...  By combining the stacked feature extraction and the weighted feature selection for D-FES, our experiment was verified to get the best performance of detection rate, 99.918% and false alarm rate, 0.012%  ...  A time series analysis by using LSTM-networks promise a good anomaly detector. However, again, the training workload still high for real-time analysis.  ... 
doi:10.1587/transinf.2019ici0001 fatcat:s7s74xh3prhn5pho55pigyl7re

Not to Cry Wolf: Distantly Supervised Multitask Learning in Critical Care [article]

Patrick Schwab, Emanuela Keller, Carl Muroi, David J. Mack, Christian Strässle, Walter Karlen
2018 arXiv   pre-print
multiple related auxiliary tasks in order to reduce the number of expensive labels required for training.  ...  We frame the problem of false alarm reduction from multivariate time series as a machine-learning task and address it with a novel multitask network architecture that utilises distant supervision through  ...  Acknowledgements This work was partially funded by the Swiss National Science Foundation (SNSF) project No. 167195 within the National Research Program (NRP) 75 "Big Data" and the Swiss Commission for  ... 
arXiv:1802.05027v2 fatcat:3p2jj2mrs5hf3fetslvhi2gdmm

Feature Extraction and Analysis of Natural Language Processing for Deep Learning English Language

Dongyang Wang, Junli Su, Hongbin Yu
2020 IEEE Access  
, this paper introduces the feature extraction method of deep learning and applies the ideas of deep learning to multi-modal feature extraction.  ...  For each mode, there is a multilayer sub-neural network with an independent structure corresponding to it. It is used to convert the features in different modes to the same-modal features.  ...  At the same time, it was found that using unsupervised data for model training can improve the discriminative ability of the model to extract features.  ... 
doi:10.1109/access.2020.2974101 fatcat:epkpxqrtyzey7e3q5xiq5jnneu

Self-supervised Autoregressive Domain Adaptation for Time Series Data [article]

Mohamed Ragab, Emadeldeen Eldele, Zhenghua Chen, Min Wu, Chee-Keong Kwoh, Xiaoli Li
2021 arXiv   pre-print
., ImageNet) for the source pretraining, which is not applicable for time-series data.  ...  Yet, these approaches may have limited performance for time series data due to the following reasons.  ...  Chen et al. presented a high-order MMD to match the high-order moments between the source and target domains [20] .  ... 
arXiv:2111.14834v1 fatcat:v6jgc4uhdfcudfvxxu4kkpwxm4

TransForensics: Image Forgery Localization with Dense Self-Attention [article]

Jing Hao and Zhixin Zhang and Shicai Yang and Di Xie and Shiliang Pu
2021 arXiv   pre-print
The former is to model global context and all pairwise interactions between local patches at different scales, while the latter is used for improving the transparency of the hidden layers and correcting  ...  types and patch sequence orders.  ...  So the discriminative features between patches can be extracted by modeling the relations between points in feature maps. In this paper, we do not split the whole image into a series of patches.  ... 
arXiv:2108.03871v1 fatcat:r2bzn53ptbhsbg4d4bo5w2zobq

SODA: Detecting COVID-19 in Chest X-rays with Semi-supervised Open Set Domain Adaptation

Jieli Zhou, Baoyu Jing, Zeya Wang, Hongyi Xin, Hanghang Tong
2021 IEEE/ACM Transactions on Computational Biology & Bioinformatics  
method, Semi-supervised Open set Domain Adversarial network (SODA).  ...  In our experiments, SODA achieves a leading classification performance compared with recent state-of-the-art models in separating COVID-19 with common pneumonia.  ...  Visualization We use t-SNE to project the high dimensional hidden features h extracted by DANN, PADA, and SODA to low dimensional space.  ... 
doi:10.1109/tcbb.2021.3066331 pmid:33729944 fatcat:c4v7unxei5cgvkvsc346d6iakm

A Centrifugal Pump Fault Diagnosis Framework Based on Supervised Contrastive Learning

Sajjad Ahmad, Zahoor Ahmad, Jong-Myon Kim
2022 Sensors  
To extract the discriminant features related to faults from the kurtogram images, we used a deep learning tool convolutional encoder (CE) with a supervised contrastive loss.  ...  First, to visualize the fault-related impulses in vibration data, we computed the kurtogram images of time series vibration sequences.  ...  In order to find the impactful and discriminant parts in vibration signals, we computed fast kurtograms of the time series vibration signals.  ... 
doi:10.3390/s22176448 pmid:36080907 pmcid:PMC9460177 fatcat:246qepfibfcmrf6pd3sk75wb5u

Aircraft Type Recognition in Remote Sensing Images: Bilinear Discriminative Extreme Learning Machine Framework

Baojun Zhao, Wei Tang, Yu Pan, Yuqi Han, Wenzheng Wang
2021 Electronics  
The bilinear pooling model uses the feature association information for feature fusion to enhance the substantial distinction of features.  ...  Furthermore, the manifold regularized ELM-AE (MRELM-AE), which can simultaneously consider the geometrical structure and discriminative information of aircraft data, is developed to extract discriminative  ...  Acknowledgments: The authors would like to thank all reviewers and editors for their constructive comments for this study. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/electronics10172046 fatcat:nl32ybgumra4bk6qs4uyirisja

Time-series classification using neural Bag-of-Features

Nikolaos Passalis, Avraam Tsantekidis, Anastasios Tefas, Juho Kanniainen, Moncef Gabbouj, Alexandros Iosifidis
2017 2017 25th European Signal Processing Conference (EUSIPCO)  
In this work, a neural generalization of the BoF model, composed of an RBF layer and an accumulation layer, is proposed as a neural layer that receives the features extracted from a time-series and gradually  ...  The proposed method can be combined with any other layer or classifier, such as fully connected layers or feature transformation layers, to form deep neural networks for time-series classification.  ...  In [8] , the BoF model is used for time-series representation and the dictionary is optimized using a discriminative loss function.  ... 
doi:10.23919/eusipco.2017.8081217 dblp:conf/eusipco/PassalisTTKGI17 fatcat:g22njf6egzbdvnfoncu7ya34ye

Human Activity Recognition as Time-Series Analysis

Hyesuk Kim, Incheol Kim
2015 Mathematical Problems in Engineering  
In addition, in order to get view-invariant, but more informative features, we extract joint angles from the subject's skeleton model and then perform the feature transformation to obtain three different  ...  In order to model effectively these high-level daily activities, we utilized a multiclass HCRF model, which is a kind of probabilistic graphical models.  ...  In this work, activities are modeled with Hidden Markov Model (HMM). The HMM is a widely used probabilistic graphical model to process a time-series data.  ... 
doi:10.1155/2015/676090 fatcat:jzriykrx5ffmbj75zdevxwg2vm

Conditional restricted Boltzmann machine as a generative model for body-worn sensor signals

Erkan Karakus, Hatice Kose
2021 IET Signal Processing  
Sensor-based human activity classification requires time and frequency domain feature extraction techniques.  ...  Conditional restricted Boltzmann machines (CRBMs) is an extension to RBM, which can capture temporal information in time-series signals and can be deployed as a generative model in classification.  ...  Time series feature extraction and feature learning is a key step used in unsupervised, semi-supervised and supervised applications [9, 10] .  ... 
doi:10.1049/iet-spr.2020.0154 fatcat:uemmr27ztvhtzg3ywpthl35s3u

Supervised Machine Learning Model for High Dimensional Gene Data in Colon Cancer Detection

Huaming Chen, Hong Zhao, Jun Shen, Rui Zhou, Qingguo Zhou
2015 2015 IEEE International Congress on Big Data  
In the supervised model, we demonstrate a shallow neural network model with a batch of parameters, and narrow its computational process into several positive parts, which process smoothly for a better  ...  In the supervised model, we demonstrate a shallow neural network model with a batch of parameters, and narrow its computational process into several positive parts, which process smoothly for a better  ...  In future we will conduct this supervised machine learning model in more datasets, as well as time series prediction and function fitting areas.  ... 
doi:10.1109/bigdatacongress.2015.28 dblp:conf/bigdata/ChenZS0Z15 fatcat:x7bwoqgnsnba7k3pn52rb6ska4

Real-Time Medical Electronic Data Mining Based Hierarchical Attention Mechanism

Yi Mao, Yun Li, Yixin Chen
2020 ICIC Express Letters  
time series.  ...  In this paper, we recur to hierarchical attention and encoder-to-decoder based model to automatically learn features from medical records of time series of vital sign, categorical features which include  ...  We propose a framework to effectively train deep architectures to learn hidden discriminant features from the original time series in an end-to-end manner.  ... 
doi:10.24507/icicel.14.12.1155 fatcat:hqbycbrvknbltmihden2hgl6v4

A Supervised Time Series Feature Extraction Technique Using DCT and DWT

Iyad Batal, Milos Hauskrecht
2009 2009 International Conference on Machine Learning and Applications  
Time series data are usually characterized by a high space dimensionality and a very strong correlation among features.  ...  Then, it performs supervised feature selection/reduction by selecting only the most discriminative set of coefficients to represent the data.  ...  In this paper, we study a supervised spectral feature extraction techniques for time series classification problems.  ... 
doi:10.1109/icmla.2009.13 dblp:conf/icmla/BatalH09 fatcat:6rrova6635e5xeuirxi43zemgq
« Previous Showing results 1 — 15 out of 26,133 results