Most gear fault diagnosis(GFD)approaches su er from ine ciency when facing with multiple varying working conditions at the same time.In this paper,a non-negative matrix factorization(NMF)-theoretic co-clustering strat...Most gear fault diagnosis(GFD)approaches su er from ine ciency when facing with multiple varying working conditions at the same time.In this paper,a non-negative matrix factorization(NMF)-theoretic co-clustering strategy is proposed specially to classify more than one task at the same time using the high dimension matrix,aiming to o er a fast multi-tasking solution.The short-time Fourier transform(STFT)is first used to obtain the time-frequency features from the gear vibration signal.Then,the optimal clustering numbers are estimated using the Bayesian information criterion(BIC)theory,which possesses the simultaneous assessment capability,compared with traditional validity indexes.Subsequently,the classical/modified NMF-based co-clustering methods are carried out to obtain the classification results in both row and column tasks.Finally,the parameters involved in BIC and NMF algorithms are determined using the gradient ascent(GA)strategy in order to achieve reliable diagnostic results.The Spectra Quest’s Drivetrain Dynamics Simulator gear data sets were analyzed to verify the e ectiveness of the proposed approach.展开更多
[Objectives] To use co-clustering analysis and visualization method to analyze the research on Siraitiae Fructus in recent ten years,to know the hot spots and trend of research. [Methods] Relevant research results abo...[Objectives] To use co-clustering analysis and visualization method to analyze the research on Siraitiae Fructus in recent ten years,to know the hot spots and trend of research. [Methods] Relevant research results about S. Fructus in CNKI from January of 2007 to December of 2016 were retrieved by computers,and the retrieval time was February 20,2017. BICOMB,Net Draw,g CLUTO and SPSS19. 0 software were used to conduct co-clustering analysis and visualization analysis for included articles. Keywords were analyzed,and social network graph,visualization matrix,peak image and multidimensional scaling analysis map were drawn. Correlation among high-frequency key words were analyzed. [Results] Totally 723 articles were included,among which 70 articles were issued during 2012-2016; 76 key words were obtained by key word co-occurrence network map,among which mogroside,MOG,extraction process,tissue culture,cultivation technology,varieties,growth and development were in the core position; visualization and the peak image showed that the topics in this research field could be divided into 6 categories; research hotspot dynamic evolution showed that S. Fructus flower,beverage,total flavonoids,gene expression,gene cloning,enzyme,apoptosis,and S. Fructus seed oil would be the hot spots of further study. [Conclusions]This study reveals that the research on S. Fructus in the recent ten years is becoming mature,and expanding to deep level. This study can be promoted to discipline development evaluation of TCM research field.展开更多
Ransomware has emerged as a critical cybersecurity threat,characterized by its ability to encrypt user data or lock devices,demanding ransom for their release.Traditional ransomware detection methods face limitations ...Ransomware has emerged as a critical cybersecurity threat,characterized by its ability to encrypt user data or lock devices,demanding ransom for their release.Traditional ransomware detection methods face limitations due to their assumption of similar data distributions between training and testing phases,rendering them less effective against evolving ransomware families.This paper introduces TLERAD(Transfer Learning for Enhanced Ransomware Attack Detection),a novel approach that leverages unsupervised transfer learning and co-clustering techniques to bridge the gap between source and target domains,enabling robust detection of both known and unknown ransomware variants.The proposed method achieves high detection accuracy,with an AUC of 0.98 for known ransomware and 0.93 for unknown ransomware,significantly outperforming baseline methods.Comprehensive experiments demonstrate TLERAD’s effectiveness in real-world scenarios,highlighting its adapt-ability to the rapidly evolving ransomware landscape.The paper also discusses future directions for enhancing TLERAD,including real-time adaptation,integration with lightweight and post-quantum cryptography,and the incorporation of explainable AI techniques.展开更多
为了同时对数据对象和特征进行聚类分析以提高聚类准确率,在模糊紧致性和分离性算法(fuzzy compactness and separation,FCS)基础上,提出一种结合类内紧致性和类间分离性的模糊联合聚类算法(fuzzy compactness and separation co-cluste...为了同时对数据对象和特征进行聚类分析以提高聚类准确率,在模糊紧致性和分离性算法(fuzzy compactness and separation,FCS)基础上,提出一种结合类内紧致性和类间分离性的模糊联合聚类算法(fuzzy compactness and separation co-clustering,FCSCC)。该算法在FCS的基础上增加了对特征维度的隶属度关系与熵最大化原理,能够在数据对象和特征2个维度上同时聚类。为验证该算法的有效性,另选择了3种算法在5个数据集上进行了对比实验,结果表明,FCSCC算法的聚类准确率高于其他3种算法。展开更多
Large amounts of data accumulated in ecology and related environmental sciences arouses urgent need to explore useful patterns and information in it.Here we propose coclustering-based methods and a temperatures-photop...Large amounts of data accumulated in ecology and related environmental sciences arouses urgent need to explore useful patterns and information in it.Here we propose coclustering-based methods and a temperatures-photoperiod driven phenological model to explore spatio-temporal differentiation in long-term spring phenology in China.First,we created the first bloom date(FBD)dataset in China from 1979 to 2018 using the extended spring indices and China Meteorological Forcing Dataset.Then we analyzed the dataset using Bregman block average co-clustering algorithm with I-divergence(BBAC_I)and kmeans algorithm.Such analysis delineated the spatially-continuous phenoregions in China for the first time.Results showed three spatial patterns of FBD in China and their temporal dynamics for 40 years(1979–2018).More specifically,overall late spring onsets occur in 1979–1996,in which areas located in Jiangxi,northern Xinjiang and middle Inner Mongolia experienced constant changing spring onsets.Overall increasingly earlier spring onsets occur in 1997–2012,in which areas located in Fujian,Hunan and eastern Heilongjiang experienced the most variable spring onsets.Stable early spring onsets over China occur after 2012.Results also showed 15 temporal patterns of spring phenology over the study period and their spatial delineation in China.More specifically,most areas in China have the same FBD category for 40 years while northern Guizhou,Hunan and southern Hubei have the same category in 1979–1997 and then fluctuate between different categories.Finally,our results have certain directive significance on the design of existing observational sites in Chinese Phenological Network.展开更多
基金Supported by National Natural Science Foundation of China(Grant No.51575102)Jiangsu Postgraduate Research Innovation Program(Grant No.KYCX18_0075).
文摘Most gear fault diagnosis(GFD)approaches su er from ine ciency when facing with multiple varying working conditions at the same time.In this paper,a non-negative matrix factorization(NMF)-theoretic co-clustering strategy is proposed specially to classify more than one task at the same time using the high dimension matrix,aiming to o er a fast multi-tasking solution.The short-time Fourier transform(STFT)is first used to obtain the time-frequency features from the gear vibration signal.Then,the optimal clustering numbers are estimated using the Bayesian information criterion(BIC)theory,which possesses the simultaneous assessment capability,compared with traditional validity indexes.Subsequently,the classical/modified NMF-based co-clustering methods are carried out to obtain the classification results in both row and column tasks.Finally,the parameters involved in BIC and NMF algorithms are determined using the gradient ascent(GA)strategy in order to achieve reliable diagnostic results.The Spectra Quest’s Drivetrain Dynamics Simulator gear data sets were analyzed to verify the e ectiveness of the proposed approach.
基金Supported by Guangxi Major Scientific Research and Technological Development Project(Gui Ke Zhong 1355001-4,14124002-11)
文摘[Objectives] To use co-clustering analysis and visualization method to analyze the research on Siraitiae Fructus in recent ten years,to know the hot spots and trend of research. [Methods] Relevant research results about S. Fructus in CNKI from January of 2007 to December of 2016 were retrieved by computers,and the retrieval time was February 20,2017. BICOMB,Net Draw,g CLUTO and SPSS19. 0 software were used to conduct co-clustering analysis and visualization analysis for included articles. Keywords were analyzed,and social network graph,visualization matrix,peak image and multidimensional scaling analysis map were drawn. Correlation among high-frequency key words were analyzed. [Results] Totally 723 articles were included,among which 70 articles were issued during 2012-2016; 76 key words were obtained by key word co-occurrence network map,among which mogroside,MOG,extraction process,tissue culture,cultivation technology,varieties,growth and development were in the core position; visualization and the peak image showed that the topics in this research field could be divided into 6 categories; research hotspot dynamic evolution showed that S. Fructus flower,beverage,total flavonoids,gene expression,gene cloning,enzyme,apoptosis,and S. Fructus seed oil would be the hot spots of further study. [Conclusions]This study reveals that the research on S. Fructus in the recent ten years is becoming mature,and expanding to deep level. This study can be promoted to discipline development evaluation of TCM research field.
文摘Ransomware has emerged as a critical cybersecurity threat,characterized by its ability to encrypt user data or lock devices,demanding ransom for their release.Traditional ransomware detection methods face limitations due to their assumption of similar data distributions between training and testing phases,rendering them less effective against evolving ransomware families.This paper introduces TLERAD(Transfer Learning for Enhanced Ransomware Attack Detection),a novel approach that leverages unsupervised transfer learning and co-clustering techniques to bridge the gap between source and target domains,enabling robust detection of both known and unknown ransomware variants.The proposed method achieves high detection accuracy,with an AUC of 0.98 for known ransomware and 0.93 for unknown ransomware,significantly outperforming baseline methods.Comprehensive experiments demonstrate TLERAD’s effectiveness in real-world scenarios,highlighting its adapt-ability to the rapidly evolving ransomware landscape.The paper also discusses future directions for enhancing TLERAD,including real-time adaptation,integration with lightweight and post-quantum cryptography,and the incorporation of explainable AI techniques.
文摘为了同时对数据对象和特征进行聚类分析以提高聚类准确率,在模糊紧致性和分离性算法(fuzzy compactness and separation,FCS)基础上,提出一种结合类内紧致性和类间分离性的模糊联合聚类算法(fuzzy compactness and separation co-clustering,FCSCC)。该算法在FCS的基础上增加了对特征维度的隶属度关系与熵最大化原理,能够在数据对象和特征2个维度上同时聚类。为验证该算法的有效性,另选择了3种算法在5个数据集上进行了对比实验,结果表明,FCSCC算法的聚类准确率高于其他3种算法。
基金supported by the National Key R&D Program of China(Grant No.2019YFA0606901)the National Natural Science Foundation of China(Grant No.41901317)the China Postdoctoral Science Foundation(Grant No.2018M641246)。
文摘Large amounts of data accumulated in ecology and related environmental sciences arouses urgent need to explore useful patterns and information in it.Here we propose coclustering-based methods and a temperatures-photoperiod driven phenological model to explore spatio-temporal differentiation in long-term spring phenology in China.First,we created the first bloom date(FBD)dataset in China from 1979 to 2018 using the extended spring indices and China Meteorological Forcing Dataset.Then we analyzed the dataset using Bregman block average co-clustering algorithm with I-divergence(BBAC_I)and kmeans algorithm.Such analysis delineated the spatially-continuous phenoregions in China for the first time.Results showed three spatial patterns of FBD in China and their temporal dynamics for 40 years(1979–2018).More specifically,overall late spring onsets occur in 1979–1996,in which areas located in Jiangxi,northern Xinjiang and middle Inner Mongolia experienced constant changing spring onsets.Overall increasingly earlier spring onsets occur in 1997–2012,in which areas located in Fujian,Hunan and eastern Heilongjiang experienced the most variable spring onsets.Stable early spring onsets over China occur after 2012.Results also showed 15 temporal patterns of spring phenology over the study period and their spatial delineation in China.More specifically,most areas in China have the same FBD category for 40 years while northern Guizhou,Hunan and southern Hubei have the same category in 1979–1997 and then fluctuate between different categories.Finally,our results have certain directive significance on the design of existing observational sites in Chinese Phenological Network.
