Fault Diagnosis of Industrial Process Based on FDKICA-PCA by ZHANG Jing, ZHU Fei-fei, LIU Jia-xing, WANG Jiang-tao

“…Because the dynamic characteristics of autocorrelation and lag correlation in production process are neglected in fault diagnosis,Kernel Independent Component AnalysisPrincipal Component Analysis (KICAPCA) is very poor in detecting small and gradual faults because of lacking available variable contribution analysis.In this paper, a dynamic kernel independent component analysis (KICAPCA) fault diagnosis method based on wavelet packet filtering is proposed.This method integrates wavelet packet filtering theory and AR model prediction data characteristics into KICAPCA to extract the feature information of process variable autocorrelation and lagrelated .In this paper, KICAPCA algorithm is used to extract the independent components and principal components of process variables to determine the control limits of three monitoring indicators T2, SPE,I2.Nonlinear contribution graph is used for fault diagnosis, and the advantage of FDKICAPCA method is verified by simulation results of Tennessee process.…”
Get full text

Real-Time Acoustic Measurement System for Cutting-Tool Analysis During Stainless Steel Machining by Tom Salm, Kourosh Tatar, José Chilo

“…Using the TreeBagger machine-learning algorithm, the system accurately predicts tool wear, detecting both gradual and abrupt wear patterns. …”
Get full text

Determination of Spatiotemporal Gait Parameters Using a Smartphone’s IMU in the Pocket: Threshold-Based and Deep Learning Approaches by Seunghee Lee, Changeon Park, Eunho Ha, Jiseon Hong, Sung Hoon Kim, Youngho Kim

“…This study proposes a hybrid approach combining threshold-based algorithm and deep learning to detect four major gait events—initial contact (IC), toe-off (TO), opposite initial contact (OIC), and opposite toe-off (OTO)—using only a smartphone’s built-in inertial sensor placed in the user’s pocket. …”
Get full text

A parallel CNN architecture for sport activity recognition based on minimal movement data by Huipeng Zhao

“…Every signal representation is utilized as an input for a Separated convolutional model, which constructs the motion features using the sports motion information. When the two sets of motion pointsets from each CNN are merged, the situation becomes more balanced, and the Random Forest classification model is able to identify the type of sports activity by detecting and classifying the features. …”
Get full text

Machine Learning-Driven Prediction of Vitamin D Deficiency Severity with Hybrid Optimization by Usharani Bhimavarapu, Gopi Battineni, Nalini Chintalapudi

“…The improved whale optimization (IWOA) algorithm was used for feature selection, which optimized weight functions to improve prediction accuracy. …”
Get full text

Modeling and countermeasures of a social network-based botnet with strong destroy-resistance by Tao YIN, Shi-cong LI, Yu-peng TUO, Yong-zheng ZHANG

“…To defeat botnets and ensure cyberspace security,a novel social network-based botnet with strong destroy-resistance (DR-SNbot),as well as its corresponding countermeasure,was proposed.DR-SNbot constructed command and control servers (C＆C-Servers) based on social network.Each C＆C-Server corresponded to a unique pseudo-random nickname.The botmaster issues commanded by hiding them in diaries using information hiding techniques,and then a novel C＆C channel was established.When different proportions of C＆C-Servers were invalid,DR-SNbot would send out different levels of alarms to inform attackers to construct new C＆C-Servers.Then,DR-SNbot could automatically repair C＆C communication to ensure its strong destroy-resistance.Under the experimental settings,DR-SNbot could resume the C＆C communication in a short period of time to keep 100% of the control rate even if all the current C＆C-Servers were invalid.Finally,a botnet nickname detecting method was proposed based on the difference of lexical features of legal nicknames and pseudo-random nicknames.Experimental results show that the proposed method can effectively (precision:96.88%,recall:93%) detect pseudo-random nicknames generated by social network-based botnets with customized algorithms.…”
Get full text

SmoothDectector: A Smoothed Dirichlet Multimodal Approach for Combating Fake News on Social Media by Akinlolu Oluwabusayo Ojo, Fatma Najar, Nuha Zamzami, Hanen T. Himdi, Nizar Bouguila

“…Our results underscore the potential of integrating a probabilistic algorithm with a deep neural network to address the challenges of fake news detection in a multimodal setting. …”
Get full text

Diagnosis of Sleep Apnea Hypopnea Syndrome Using Fusion of Micro-motion Signals from Millimeter-wave Radar and Pulse Wave Data by Xiang ZHAO, Wei WANG, Chenyang LI, Jian GUAN, Gang LI

“…This study used a radar and pulse wave data preprocessing algorithm to extract time-frequency information and artificial features from the signals, balancing the accuracy and robustness of sleep-breathing abnormality event detection Additionally, a deep neural network was designed to fuse the two types of signals for precise identification of sleep apnea and hypopnea events, and to estimate the Apnea-Hypopnea Index (AHI) for quantitative assessment of sleep-breathing abnormality severity. …”
Get full text

Android malware family classification method based on the image of bytecodeConstruction of MDS matrices by Yi-min YANG, Tie-ming CHEN

“…An Android malware family classification method based on the image of bytecode was proposed accord-ing to the exponential growth of Android malware.A bytecode file of Android malware was converted to a 256-level grayscale image and texture features was extracted from the image by GIST.The random forest algorithm was ap-plied to classify the extracted features.The method by the experimental data of 14 kinds of common Android mal-ware families was verified and was compared against the DREBIN on the same dataset.The experimental results show that the proposed method has high detection precision and low false positive rate.…”
Get full text

Molten Pool Image Segmentation Based on Adaptive Multi-Scale Attention Mechanism by Yuefeng Chen, Weihua Liu, Qing Yang, Huabin Chen, Qi Jiang

“…Initially, an adaptive multi-scale channel attention mechanism is introduced and fused with a feature pyramid, forming a multi-scale attention pyramid structure that significantly enhances the network’s capability to extract intricate molten pool features. …”
Get full text

UHVDC Transmission Line Fault Identification Method Based on Generalized Regression Neural Network by XIE Jia, LIU Feng, KE Yanguo, YIN Zhen, RUAN Wei, YAO Jinming

“…A protection method for ultra-high voltage direct current transmission lines based on generalized regression neural network (GRNN) is proposed to address the issues of easy rejection and long fault detection time in ultra-high voltage direct current protection. …”
Get full text

An Anti-Interrupted-Sampling Repeater Jamming Method Based on Simulated Annealing–2-Optimization Parallel Optimization of Waveforms and Fractional Domain Extraction by Ziming Yin, Pengcheng Guo, Yunyu Wei, Sizhe Gao, Jingjing Wang, Anxiang Xue, Kuo Wang

“…Then, the received signal is subjected to the fractional Fourier transform (FrFT) and inverse transform to extract the target signal. Finally, jamming detection is conducted based on the multi-dimensional features of the pulse-compressed signal. …”
Get full text

External phantom-based validation of a deep-learning network trained for upscaling of digital low count PET data by Anja Braune, René Hosch, David Kersting, Juliane Müller, Frank Hofheinz, Ken Herrmann, Felix Nensa, Jörg Kotzerke, Robert Seifert

“…The performance of this algorithm has so far only been clinically evaluated on patient data featuring limited scan statistics and unknown actual activity concentration. …”
Get full text

A Portable Smartphone-Based 3D-Printed Biosensing Platform for Kidney Function Biomarker Quantification by Sangeeta Palekar, Sharayu Kalambe, Jayu Kalambe, Madhusudan B. Kulkarni, Manish Bhaiyya

“…The system adapts to variations in smartphones, reagent brands, and lighting conditions through an adaptive calibration algorithm, ensuring flexibility and accuracy. The sensor demonstrated good linear detection ranges for uric acid (1–30 mg/dL), creatinine (0.1–20 mg/dL), and albumin (0.1–8 g/dL), with detection limits of 1.15 mg/dL, 0.15 mg/dL, and 0.11 g/dL, respectively. …”
Get full text

Curvature estimation for point cloud 2-manifolds based on the heat kernel by Kai Wang, Xiheng Wang, Xiaoping Wang

“…As an exemplary application, we utilized the mean curvature for detecting features of point clouds.…”
Get full text

Analysis of Acoustic Signals of Footsteps from the Piezoelectric Sensor by Bilge Çiğdem Çiftçi, Gamze Kaya, Mustafa Kurt

“…The original signal was pre-processed using filtering systems and analyzed by the fast Fourier transform and power spectral density methods to extract descriptive spectral features of the signal. This preliminary study proposed a method as a sensor based piezoelectric security system to detect the acoustic signals that can indicate possible dangers to the safety of people or property. …”
Get full text

Potential distribution of native freshwater fish in Tabasco, Mexico by Paula Andrea Castillo-Torres, Enrique Martínez-Meyer, Fernando Cordova Tapia, Luis Zambrano

“…We used occurrence records of 41 native fish species and 22 interpolated surfaces that represent topographic features and annual, seasonal and extreme trends of temperature and precipitation to generate niche-based potential geographic distribution maps using the GARP and MaxEnt algorithms. …”
Get full text

A segmentation method for lung nodule image sequences based on superpixels and density-based spatial clustering of applications with noise. by Wei Zhang, Xiaolong Zhang, Juanjuan Zhao, Yan Qiang, Qi Tian, Xiaoxian Tang

“…Moreover, by fitting the distance and detecting the change in slope, an accurate clustering threshold is obtained. …”
Get full text

Development of a deep learning method to identify acute ischaemic stroke lesions on brain CT by Emanuele Trucco, Joanna M Wardlaw, Wenwen Li, Grant Mair, Amos Storkey, Alessandro Fontanella, Antreas Antoniou, Eleanor Platt, Paul Armitage

“…We aimed to develop a deep learning (DL) method using CT brain scans that were labelled but not annotated for the presence of ischaemic lesions.Methods We designed a convolutional neural network-based DL algorithm to detect ischaemic lesions on CT. Our algorithm was trained using routinely acquired CT brain scans collected for a large multicentre international trial. …”
Get full text

Artificial intelligence analysis in cyber domain: A review by Liguo Zhao, Derong Zhu, Wasswa Shafik, S Mojtaba Matinkhah, Zubair Ahmad, Lule Sharif, Alisa Craig

“…An in-depth learning algorithm is utilized for training a neural network for detecting suspicious user activities. …”
Get full text