克隆合作猪神经营养因子3基因(Neurotrophin3,NTF3)的编码区序列(Coding region sequence,CDS),预测分析其结构与功能,并检测基因在不同组织中的表达量,构建组织表达谱。通过PCR扩增、Sanger测序等方法获得合作猪NTF3基因CDS区序列,使...克隆合作猪神经营养因子3基因(Neurotrophin3,NTF3)的编码区序列(Coding region sequence,CDS),预测分析其结构与功能,并检测基因在不同组织中的表达量,构建组织表达谱。通过PCR扩增、Sanger测序等方法获得合作猪NTF3基因CDS区序列,使用生物信息学方法分析其结构,并使用RT-qPCR技术检测其在不同组织中的表达水平。结果显示,合作猪NTF3基因CDS区长774 bp,编码257个氨基酸;等电点(pI)为9.46,属于碱性蛋白;合作猪与猪、牛亲缘关系较近;二级结构主要为α-螺旋和无规则卷曲,三级结构主要由β-折叠、α-螺旋等组成;整条链中亲水性氨基酸残基数量多于疏水性氨基酸残基数量,是亲水性蛋白;无跨膜区域,是非跨膜蛋白;在37~58位氨基酸残基有一个线圈结构域、64~79位有一个低复杂度结构域和144~249位有一个神经生长因子结构域;存在信号肽,推测是分泌蛋白;NTF3基因在合作猪脾脏的表达量最高,回肠表达量最低。展开更多
Dear Editor,This letter presents a novel latent factorization model for high dimensional and incomplete (HDI) tensor, namely the neural Tucker factorization (Neu Tuc F), which is a generic neural network-based latent-...Dear Editor,This letter presents a novel latent factorization model for high dimensional and incomplete (HDI) tensor, namely the neural Tucker factorization (Neu Tuc F), which is a generic neural network-based latent-factorization-of-tensors model under the Tucker decomposition framework.展开更多
Substantial effects of photochemical reaction losses of volatile organic compounds(VOCs)on factor profiles can be investigated by comparing the differences between daytime and nighttime dispersion-normalized VOC data ...Substantial effects of photochemical reaction losses of volatile organic compounds(VOCs)on factor profiles can be investigated by comparing the differences between daytime and nighttime dispersion-normalized VOC data resolved profiles.Hourly speciated VOC data measured in Shijiazhuang,China from May to September 2021 were used to conduct study.The mean VOC concentration in the daytime and at nighttime were 32.8 and 36.0 ppbv,respectively.Alkanes and aromatics concentrations in the daytime(12.9 and 3.08 ppbv)were lower than nighttime(15.5 and 3.63 ppbv),whereas that of alkenes showed the opposite tendency.The concentration differences between daytime and nighttime for alkynes and halogenated hydrocarbonswere uniformly small.The reactivities of the dominant species in factor profiles for gasoline emissions,natural gas and diesel vehicles,and liquefied petroleum gas were relatively low and their profiles were less affected by photochemical losses.Photochemical losses produced a substantial impact on the profiles of solvent use,petrochemical industry emissions,combustion sources,and biogenic emissions where the dominant species in these factor profiles had high reactivities.Although the profile of biogenic emissions was substantially affected by photochemical loss of isoprene,the low emissions at nighttime also had an important impact on its profile.Chemical losses of highly active VOC species substantially reduced their concentrations in apportioned factor profiles.This study results were consistent with the analytical results obtained through initial concentration estimation,suggesting that the initial concentration estimation could be the most effective currently availablemethod for the source analyses of active VOCs although with uncertainty.展开更多
CircRNAs,widely found throughout the human bodies,play a crucial role in regulating various biological processes and are closely linked to complex human diseases.Investigating potential associations between circRNAs a...CircRNAs,widely found throughout the human bodies,play a crucial role in regulating various biological processes and are closely linked to complex human diseases.Investigating potential associations between circRNAs and diseases can enhance our understanding of diseases and provide new strategies and tools for early diagnosis,treatment,and disease prevention.However,existing models have limitations in accurately capturing similarities,handling the sparse and noise attributes of association networks,and fully leveraging bioinformatical aspects from multiple viewpoints.To address these issues,this study introduces a new non-negative matrix factorization-based framework called NMFMSN.First,we incorporate circRNA sequence data and disease semantic information to compute circRNA and disease similarity,respectively.Given the sparse known associations between circRNAs and diseases,we reconstruct the network to complete more associations by imputing missing links based on neighboring circRNA and disease interactions.Finally,we integrate these two similarity networks into a non-negative matrix factorization framework to identify potential circRNA-disease associations.Upon conducting 5-fold cross-validation and leave-one-out cross-validation,the AUC values for NMFMSN reach 0.9712 and 0.9768,respectively,outperforming the currently most advanced models.Case studies on lung cancer and hepatocellular carcinoma show that NMFMSN is a good way to predict new associations between circRNAs and diseases.展开更多
In this paper,a nonlinear control approach for an unstable networked plant in the presence of actuator and sensor limitations using robust right coprime factorization is proposed.The actuator is limited by upper and l...In this paper,a nonlinear control approach for an unstable networked plant in the presence of actuator and sensor limitations using robust right coprime factorization is proposed.The actuator is limited by upper and lower constraints and the sensor in the feedback loop is subjected to network-induced unknown time-varying delay and noise.With this nonlinear control method,we first employ right coprime factorization based on isomorphism and operator theory to factorize the plant,so that bounded input bounded output(BIBO)stability can be guaranteed.Next,continuous-time generalized predictive control(CGPC)is utilized for the unstable operator of the right coprime factorized plant to guarantee inner stability and enables the closed-loop dynamics of the system with predictive characteristics.Meanwhile,a second-Do F(degrees of freedom)switched controller that satisfies a perturbed Bezout identity and a robustness condition is designed.By using the CGPC controller that possesses predictive behavior and the second-Do F switched stabilizer,the overall stability of the plant subjected to actuator limitations is guaranteed.To address sensor limitations that exist in networked plants in the form of delay and noise which often cause system performance degradation,we implement an identity operator definition in the feedback loop to compensate for these adverse effects.Further,a pre-operator is designed to ensure that the plant output tracks the reference input.Finally,the effectiveness of the proposed design scheme is demonstrated by simulations.展开更多
Fine particulatematter(PM_(2.5))samples were collected in two neighboring cities,Beijing and Baoding,China.High-concentration events of PM_(2.5) in which the average mass concentration exceeded 75μg/m^(3) were freque...Fine particulatematter(PM_(2.5))samples were collected in two neighboring cities,Beijing and Baoding,China.High-concentration events of PM_(2.5) in which the average mass concentration exceeded 75μg/m^(3) were frequently observed during the heating season.Dispersion Normalized Positive Matrix Factorization was applied for the source apportionment of PM_(2.5) as minimize the dilution effects of meteorology and better reflect the source strengths in these two cities.Secondary nitrate had the highest contribution for Beijing(37.3%),and residential heating/biomass burning was the largest for Baoding(27.1%).Secondary nitrate,mobile,biomass burning,district heating,oil combustion,aged sea salt sources showed significant differences between the heating and non-heating seasons in Beijing for same period(2019.01.10–2019.08.22)(Mann-Whitney Rank Sum Test P<0.05).In case of Baoding,soil,residential heating/biomass burning,incinerator,coal combustion,oil combustion sources showed significant differences.The results of Pearson correlation analysis for the common sources between the two cities showed that long-range transported sources and some sources with seasonal patterns such as oil combustion and soil had high correlation coefficients.Conditional Bivariate Probability Function(CBPF)was used to identify the inflow directions for the sources,and joint-PSCF(Potential Source Contribution Function)was performed to determine the common potential source areas for sources affecting both cities.These models facilitated a more precise verification of city-specific influences on PM_(2.5) sources.The results of this study will aid in prioritizing air pollution mitigation strategies during the heating season and strengthening air quality management to reduce the impact of downwind neighboring cities.展开更多
Dear Editor,This letter presents a latent-factorization-of-tensors(LFT)-incorporated battery cycle life prediction framework.Data-driven prognosis and health management(PHM)for battery pack(BP)can boost the safety and...Dear Editor,This letter presents a latent-factorization-of-tensors(LFT)-incorporated battery cycle life prediction framework.Data-driven prognosis and health management(PHM)for battery pack(BP)can boost the safety and sustainability of a battery management system(BMS),which relies heavily on the quality of the measured BP data like the voltage(V),current(I),and temperature(T).展开更多
This study aimed to investigate the pollution characteristics, source apportionment, and health risks associated with trace metal(loid)s(TMs) in the major agricultural producing areas in Chongqing, China. We analyzed ...This study aimed to investigate the pollution characteristics, source apportionment, and health risks associated with trace metal(loid)s(TMs) in the major agricultural producing areas in Chongqing, China. We analyzed the source apportionment and assessed the health risk of TMs in agricultural soils by using positive matrix factorization(PMF) model and health risk assessment(HRA) model based on Monte Carlo simulation. Meanwhile, we combined PMF and HRA models to explore the health risks of TMs in agricultural soils by different pollution sources to determine the priority control factors. Results showed that the average contents of cadmium(Cd), arsenic (As), lead(Pb), chromium(Cr), copper(Cu), nickel(Ni), and zinc(Zn) in the soil were found to be 0.26, 5.93, 27.14, 61.32, 23.81, 32.45, and 78.65 mg/kg, respectively. Spatial analysis and source apportionment analysis revealed that urban and industrial sources, agricultural sources, and natural sources accounted for 33.0%, 27.7%, and 39.3% of TM accumulation in the soil, respectively. In the HRA model based on Monte Carlo simulation, noncarcinogenic risks were deemed negligible(hazard index <1), the carcinogenic risks were at acceptable level(10^(-6)<total carcinogenic risk ≤ 10^(-4)), with higher risks observed for children compared to adults. The relationship between TMs, their sources, and health risks indicated that urban and industrial sources were primarily associated with As, contributing to 75.1% of carcinogenic risks and 55.7% of non-carcinogenic risks, making them the primary control factors. Meanwhile, agricultural sources were primarily linked to Cd and Pb, contributing to 13.1% of carcinogenic risks and 21.8% of non-carcinogenic risks, designating them as secondary control factors.展开更多
Contrastive learning is a significant research direction in the field of deep learning.However,existing data augmentation methods often lead to issues such as semantic drift in generated views while the complexity of ...Contrastive learning is a significant research direction in the field of deep learning.However,existing data augmentation methods often lead to issues such as semantic drift in generated views while the complexity of model pre-training limits further improvement in the performance of existing methods.To address these challenges,we propose the Efficient Clustering Network based on Matrix Factorization(ECN-MF).Specifically,we design a batched low-rank Singular Value Decomposition(SVD)algorithm for data augmentation to eliminate redundant information and uncover major patterns of variation and key information in the data.Additionally,we design a Mutual Information-Enhanced Clustering Module(MI-ECM)to accelerate the training process by leveraging a simple architecture to bring samples from the same cluster closer while pushing samples from other clusters apart.Extensive experiments on six datasets demonstrate that ECN-MF exhibits more effective performance compared to state-of-the-art algorithms.展开更多
The proliferation of internet communication channels has increased telecom fraud,causing billions of euros in losses for customers and the industry each year.Fraudsters constantly find new ways to engage in illegal ac...The proliferation of internet communication channels has increased telecom fraud,causing billions of euros in losses for customers and the industry each year.Fraudsters constantly find new ways to engage in illegal activity on the network.To reduce these losses,a new fraud detection approach is required.Telecom fraud detection involves identifying a small number of fraudulent calls from a vast amount of call traffic.Developing an effective strategy to combat fraud has become challenging.Although much effort has been made to detect fraud,most existing methods are designed for batch processing,not real-time detection.To solve this problem,we propose an online fraud detection model using a Neural Factorization Autoencoder(NFA),which analyzes customer calling patterns to detect fraudulent calls.The model employs Neural Factorization Machines(NFM)and an Autoencoder(AE)to model calling patterns and a memory module to adapt to changing customer behaviour.We evaluate our approach on a large dataset of real-world call detail records and compare it with several state-of-the-art methods.Our results show that our approach outperforms the baselines,with an AUC of 91.06%,a TPR of 91.89%,an FPR of 14.76%,and an F1-score of 95.45%.These results demonstrate the effectiveness of our approach in detecting fraud in real-time and suggest that it can be a valuable tool for preventing fraud in telecommunications networks.展开更多
One of the most important problems in complex networks is to identify the influential vertices for understanding and controlling of information diffusion and disease spreading.Most of the current centrality algorithms...One of the most important problems in complex networks is to identify the influential vertices for understanding and controlling of information diffusion and disease spreading.Most of the current centrality algorithms focus on single feature or manually extract the attributes,which occasionally results in the failure to fully capture the vertex’s importance.A new vertex centrality approach based on symmetric nonnegative matrix factorization(SNMF),called VCSNMF,is proposed in this paper.For highlight the characteristics of a network,the adjacency matrix and the degree matrix are fused to represent original data of the network via a weighted linear combination.First,SNMF automatically extracts the latent characteristics of vertices by factorizing the established original data matrix.Then we prove that each vertex’s composite feature which is constructed with one-dimensional factor matrix can be approximated as the term of eigenvector associated with the spectral radius of the network,otherwise obtained by the factor matrix on the hyperspace.Finally,VCSNMF integrates the composite feature and the topological structure to evaluate the performance of vertices.To verify the effectiveness of the VCSNMF criterion,eight existing centrality approaches are used as comparison measures to rank influential vertices in ten real-world networks.The experimental results assert the superiority of the method.展开更多
文摘克隆合作猪神经营养因子3基因(Neurotrophin3,NTF3)的编码区序列(Coding region sequence,CDS),预测分析其结构与功能,并检测基因在不同组织中的表达量,构建组织表达谱。通过PCR扩增、Sanger测序等方法获得合作猪NTF3基因CDS区序列,使用生物信息学方法分析其结构,并使用RT-qPCR技术检测其在不同组织中的表达水平。结果显示,合作猪NTF3基因CDS区长774 bp,编码257个氨基酸;等电点(pI)为9.46,属于碱性蛋白;合作猪与猪、牛亲缘关系较近;二级结构主要为α-螺旋和无规则卷曲,三级结构主要由β-折叠、α-螺旋等组成;整条链中亲水性氨基酸残基数量多于疏水性氨基酸残基数量,是亲水性蛋白;无跨膜区域,是非跨膜蛋白;在37~58位氨基酸残基有一个线圈结构域、64~79位有一个低复杂度结构域和144~249位有一个神经生长因子结构域;存在信号肽,推测是分泌蛋白;NTF3基因在合作猪脾脏的表达量最高,回肠表达量最低。
基金supported by the National Natural Science Foundation of China(62272078)Chongqing Natural Science Foundation(CSTB2023NSCQ-LZX0069)the Science and Technology Research Program of Chongqing Municipal Education Commission(KJQN202300210)
文摘Dear Editor,This letter presents a novel latent factorization model for high dimensional and incomplete (HDI) tensor, namely the neural Tucker factorization (Neu Tuc F), which is a generic neural network-based latent-factorization-of-tensors model under the Tucker decomposition framework.
基金supported by the National Key R&D Program of China(No.2023YFC3705801)the National Natural Science Foundation of China(No.42177085).
文摘Substantial effects of photochemical reaction losses of volatile organic compounds(VOCs)on factor profiles can be investigated by comparing the differences between daytime and nighttime dispersion-normalized VOC data resolved profiles.Hourly speciated VOC data measured in Shijiazhuang,China from May to September 2021 were used to conduct study.The mean VOC concentration in the daytime and at nighttime were 32.8 and 36.0 ppbv,respectively.Alkanes and aromatics concentrations in the daytime(12.9 and 3.08 ppbv)were lower than nighttime(15.5 and 3.63 ppbv),whereas that of alkenes showed the opposite tendency.The concentration differences between daytime and nighttime for alkynes and halogenated hydrocarbonswere uniformly small.The reactivities of the dominant species in factor profiles for gasoline emissions,natural gas and diesel vehicles,and liquefied petroleum gas were relatively low and their profiles were less affected by photochemical losses.Photochemical losses produced a substantial impact on the profiles of solvent use,petrochemical industry emissions,combustion sources,and biogenic emissions where the dominant species in these factor profiles had high reactivities.Although the profile of biogenic emissions was substantially affected by photochemical loss of isoprene,the low emissions at nighttime also had an important impact on its profile.Chemical losses of highly active VOC species substantially reduced their concentrations in apportioned factor profiles.This study results were consistent with the analytical results obtained through initial concentration estimation,suggesting that the initial concentration estimation could be the most effective currently availablemethod for the source analyses of active VOCs although with uncertainty.
基金the Gansu Province Industrial Support Plan(No.2023CYZC-25)Natural Science Foundation of Gansu Province(No.23JRRA770)the National Natural Science Foundation of China(No.62162040)。
文摘CircRNAs,widely found throughout the human bodies,play a crucial role in regulating various biological processes and are closely linked to complex human diseases.Investigating potential associations between circRNAs and diseases can enhance our understanding of diseases and provide new strategies and tools for early diagnosis,treatment,and disease prevention.However,existing models have limitations in accurately capturing similarities,handling the sparse and noise attributes of association networks,and fully leveraging bioinformatical aspects from multiple viewpoints.To address these issues,this study introduces a new non-negative matrix factorization-based framework called NMFMSN.First,we incorporate circRNA sequence data and disease semantic information to compute circRNA and disease similarity,respectively.Given the sparse known associations between circRNAs and diseases,we reconstruct the network to complete more associations by imputing missing links based on neighboring circRNA and disease interactions.Finally,we integrate these two similarity networks into a non-negative matrix factorization framework to identify potential circRNA-disease associations.Upon conducting 5-fold cross-validation and leave-one-out cross-validation,the AUC values for NMFMSN reach 0.9712 and 0.9768,respectively,outperforming the currently most advanced models.Case studies on lung cancer and hepatocellular carcinoma show that NMFMSN is a good way to predict new associations between circRNAs and diseases.
文摘In this paper,a nonlinear control approach for an unstable networked plant in the presence of actuator and sensor limitations using robust right coprime factorization is proposed.The actuator is limited by upper and lower constraints and the sensor in the feedback loop is subjected to network-induced unknown time-varying delay and noise.With this nonlinear control method,we first employ right coprime factorization based on isomorphism and operator theory to factorize the plant,so that bounded input bounded output(BIBO)stability can be guaranteed.Next,continuous-time generalized predictive control(CGPC)is utilized for the unstable operator of the right coprime factorized plant to guarantee inner stability and enables the closed-loop dynamics of the system with predictive characteristics.Meanwhile,a second-Do F(degrees of freedom)switched controller that satisfies a perturbed Bezout identity and a robustness condition is designed.By using the CGPC controller that possesses predictive behavior and the second-Do F switched stabilizer,the overall stability of the plant subjected to actuator limitations is guaranteed.To address sensor limitations that exist in networked plants in the form of delay and noise which often cause system performance degradation,we implement an identity operator definition in the feedback loop to compensate for these adverse effects.Further,a pre-operator is designed to ensure that the plant output tracks the reference input.Finally,the effectiveness of the proposed design scheme is demonstrated by simulations.
基金supported by the National Institute of Environmental Research(NIER)funded by the Ministry of Environment(No.NIER-2019-04-02-039)supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry&Technology Institute(KEITI)funded by the Ministry of Environment(MOE).
文摘Fine particulatematter(PM_(2.5))samples were collected in two neighboring cities,Beijing and Baoding,China.High-concentration events of PM_(2.5) in which the average mass concentration exceeded 75μg/m^(3) were frequently observed during the heating season.Dispersion Normalized Positive Matrix Factorization was applied for the source apportionment of PM_(2.5) as minimize the dilution effects of meteorology and better reflect the source strengths in these two cities.Secondary nitrate had the highest contribution for Beijing(37.3%),and residential heating/biomass burning was the largest for Baoding(27.1%).Secondary nitrate,mobile,biomass burning,district heating,oil combustion,aged sea salt sources showed significant differences between the heating and non-heating seasons in Beijing for same period(2019.01.10–2019.08.22)(Mann-Whitney Rank Sum Test P<0.05).In case of Baoding,soil,residential heating/biomass burning,incinerator,coal combustion,oil combustion sources showed significant differences.The results of Pearson correlation analysis for the common sources between the two cities showed that long-range transported sources and some sources with seasonal patterns such as oil combustion and soil had high correlation coefficients.Conditional Bivariate Probability Function(CBPF)was used to identify the inflow directions for the sources,and joint-PSCF(Potential Source Contribution Function)was performed to determine the common potential source areas for sources affecting both cities.These models facilitated a more precise verification of city-specific influences on PM_(2.5) sources.The results of this study will aid in prioritizing air pollution mitigation strategies during the heating season and strengthening air quality management to reduce the impact of downwind neighboring cities.
文摘Dear Editor,This letter presents a latent-factorization-of-tensors(LFT)-incorporated battery cycle life prediction framework.Data-driven prognosis and health management(PHM)for battery pack(BP)can boost the safety and sustainability of a battery management system(BMS),which relies heavily on the quality of the measured BP data like the voltage(V),current(I),and temperature(T).
基金supported by Project of Chongqing Science and Technology Bureau (cstc2022jxjl0005)。
文摘This study aimed to investigate the pollution characteristics, source apportionment, and health risks associated with trace metal(loid)s(TMs) in the major agricultural producing areas in Chongqing, China. We analyzed the source apportionment and assessed the health risk of TMs in agricultural soils by using positive matrix factorization(PMF) model and health risk assessment(HRA) model based on Monte Carlo simulation. Meanwhile, we combined PMF and HRA models to explore the health risks of TMs in agricultural soils by different pollution sources to determine the priority control factors. Results showed that the average contents of cadmium(Cd), arsenic (As), lead(Pb), chromium(Cr), copper(Cu), nickel(Ni), and zinc(Zn) in the soil were found to be 0.26, 5.93, 27.14, 61.32, 23.81, 32.45, and 78.65 mg/kg, respectively. Spatial analysis and source apportionment analysis revealed that urban and industrial sources, agricultural sources, and natural sources accounted for 33.0%, 27.7%, and 39.3% of TM accumulation in the soil, respectively. In the HRA model based on Monte Carlo simulation, noncarcinogenic risks were deemed negligible(hazard index <1), the carcinogenic risks were at acceptable level(10^(-6)<total carcinogenic risk ≤ 10^(-4)), with higher risks observed for children compared to adults. The relationship between TMs, their sources, and health risks indicated that urban and industrial sources were primarily associated with As, contributing to 75.1% of carcinogenic risks and 55.7% of non-carcinogenic risks, making them the primary control factors. Meanwhile, agricultural sources were primarily linked to Cd and Pb, contributing to 13.1% of carcinogenic risks and 21.8% of non-carcinogenic risks, designating them as secondary control factors.
基金supported by the Key Research and Development Program of Hainan Province(Grant Nos.ZDYF2023GXJS163,ZDYF2024GXJS014)National Natural Science Foundation of China(NSFC)(Grant Nos.62162022,62162024)+3 种基金the Major Science and Technology Project of Hainan Province(Grant No.ZDKJ2020012)Hainan Provincial Natural Science Foundation of China(Grant No.620MS021)Youth Foundation Project of Hainan Natural Science Foundation(621QN211)Innovative Research Project for Graduate Students in Hainan Province(Grant Nos.Qhys2023-96,Qhys2023-95).
文摘Contrastive learning is a significant research direction in the field of deep learning.However,existing data augmentation methods often lead to issues such as semantic drift in generated views while the complexity of model pre-training limits further improvement in the performance of existing methods.To address these challenges,we propose the Efficient Clustering Network based on Matrix Factorization(ECN-MF).Specifically,we design a batched low-rank Singular Value Decomposition(SVD)algorithm for data augmentation to eliminate redundant information and uncover major patterns of variation and key information in the data.Additionally,we design a Mutual Information-Enhanced Clustering Module(MI-ECM)to accelerate the training process by leveraging a simple architecture to bring samples from the same cluster closer while pushing samples from other clusters apart.Extensive experiments on six datasets demonstrate that ECN-MF exhibits more effective performance compared to state-of-the-art algorithms.
基金This research work has been conducted in cooperation with members of DETSI project supported by BPI France and Pays de Loire and Auvergne Rhone Alpes.
文摘The proliferation of internet communication channels has increased telecom fraud,causing billions of euros in losses for customers and the industry each year.Fraudsters constantly find new ways to engage in illegal activity on the network.To reduce these losses,a new fraud detection approach is required.Telecom fraud detection involves identifying a small number of fraudulent calls from a vast amount of call traffic.Developing an effective strategy to combat fraud has become challenging.Although much effort has been made to detect fraud,most existing methods are designed for batch processing,not real-time detection.To solve this problem,we propose an online fraud detection model using a Neural Factorization Autoencoder(NFA),which analyzes customer calling patterns to detect fraudulent calls.The model employs Neural Factorization Machines(NFM)and an Autoencoder(AE)to model calling patterns and a memory module to adapt to changing customer behaviour.We evaluate our approach on a large dataset of real-world call detail records and compare it with several state-of-the-art methods.Our results show that our approach outperforms the baselines,with an AUC of 91.06%,a TPR of 91.89%,an FPR of 14.76%,and an F1-score of 95.45%.These results demonstrate the effectiveness of our approach in detecting fraud in real-time and suggest that it can be a valuable tool for preventing fraud in telecommunications networks.
基金the National Natural Science Foundation of China(Nos.11361033 and 11861045)。
文摘One of the most important problems in complex networks is to identify the influential vertices for understanding and controlling of information diffusion and disease spreading.Most of the current centrality algorithms focus on single feature or manually extract the attributes,which occasionally results in the failure to fully capture the vertex’s importance.A new vertex centrality approach based on symmetric nonnegative matrix factorization(SNMF),called VCSNMF,is proposed in this paper.For highlight the characteristics of a network,the adjacency matrix and the degree matrix are fused to represent original data of the network via a weighted linear combination.First,SNMF automatically extracts the latent characteristics of vertices by factorizing the established original data matrix.Then we prove that each vertex’s composite feature which is constructed with one-dimensional factor matrix can be approximated as the term of eigenvector associated with the spectral radius of the network,otherwise obtained by the factor matrix on the hyperspace.Finally,VCSNMF integrates the composite feature and the topological structure to evaluate the performance of vertices.To verify the effectiveness of the VCSNMF criterion,eight existing centrality approaches are used as comparison measures to rank influential vertices in ten real-world networks.The experimental results assert the superiority of the method.
