Financial time series prediction,whether for classification or regression,has been a heated research topic over the last decade.While traditional machine learning algorithms have experienced mediocre results,deep lear...Financial time series prediction,whether for classification or regression,has been a heated research topic over the last decade.While traditional machine learning algorithms have experienced mediocre results,deep learning has largely contributed to the elevation of the prediction performance.Currently,the most up-to-date review of advanced machine learning techniques for financial time series prediction is still lacking,making it challenging for finance domain experts and relevant practitioners to determine which model potentially performs better,what techniques and components are involved,and how themodel can be designed and implemented.This review article provides an overview of techniques,components and frameworks for financial time series prediction,with an emphasis on state-of-the-art deep learning models in the literature from2015 to 2023,including standalonemodels like convolutional neural networks(CNN)that are capable of extracting spatial dependencies within data,and long short-term memory(LSTM)that is designed for handling temporal dependencies;and hybrid models integrating CNN,LSTM,attention mechanism(AM)and other techniques.For illustration and comparison purposes,models proposed in recent studies are mapped to relevant elements of a generalized framework comprised of input,output,feature extraction,prediction,and related processes.Among the state-of-the-artmodels,hybrid models like CNNLSTMand CNN-LSTM-AM in general have been reported superior in performance to stand-alone models like the CNN-only model.Some remaining challenges have been discussed,including non-friendliness for finance domain experts,delayed prediction,domain knowledge negligence,lack of standards,and inability of real-time and highfrequency predictions.The principal contributions of this paper are to provide a one-stop guide for both academia and industry to review,compare and summarize technologies and recent advances in this area,to facilitate smooth and informed implementation,and to highlight future research directions.展开更多
Correction to:Opto-Electronic Advances https://doi.org/10.29026/oea.2023.220154 published online 26 April 2023 After the publication of this article1,it was brought to our attention that calculations of the PeLEC devi...Correction to:Opto-Electronic Advances https://doi.org/10.29026/oea.2023.220154 published online 26 April 2023 After the publication of this article1,it was brought to our attention that calculations of the PeLEC device elec-troluminescent(EL)efficiency contained a mistake,leading to an inaccurate quantity value.The device’s maxim-um EL efficiency constitutes not‘~120 klm/W’but‘4.3 lm/W’instead.Correction details are listed below.展开更多
A new theoretical method to study super-multiperiod superlattices has been developed.The method combines the precision of the 8-band kp-method with the flexibility of the shooting method and the Monte Carlo approach.T...A new theoretical method to study super-multiperiod superlattices has been developed.The method combines the precision of the 8-band kp-method with the flexibility of the shooting method and the Monte Carlo approach.This method was applied to examine the finest quality samples of super-multiperiod Al_(0.3)Ga_(0.7)As/GaAs superlattices grown by molecular beam epitaxy.The express photoreflectance spectroscopy method was utilized to validate the proposed theoretical method.For the first time,the accurate theoretical analysis of the energy band diagram of super-multiperiod superlattices with experimental verification has been conducted.The proposed approach highly accurately determines transition peak positions and enables the calculation of the energy band diagram,transition energies,relaxation rates,and gain estimation.It has achieved a remarkably low 5%error compared to the commonly used method,which typically results in a 25%error,and allowed to recover the superlattice parameters.The retrieved intrinsic parameters of the samples aligned with XRD data and growth parameters.The proposed method also accurately predicted the escape of the second energy level for quantum well thicknesses less than 5 nm,as was observed in photoreflectance experiments.The new designs of THz light-emitting devices operating at room temperature were suggested by the developed method.展开更多
While hypoxic signaling has been shown to play a role in many cellular processes,its role in metabolism-linked extracellular matrix(ECM)organization and downstream processes of cell fate after musculoskeletal injury r...While hypoxic signaling has been shown to play a role in many cellular processes,its role in metabolism-linked extracellular matrix(ECM)organization and downstream processes of cell fate after musculoskeletal injury remains to be determined.Heterotopicossification(HO)is a debilitating condition where abnormal bone formation occurs within extra-skeletal tissues.Hypoxia andhypoxia-inducible factor 1α(HIF-1α)activation have been shown to promote HO.However,the underlying molecular mechanisms bywhich the HIF-1αpathway in mesenchymal progenitor cells(MPCs)contributes to pathologic bone formation remain to beelucidated.Here,we used a proven mouse injury-induced HO model to investigate the role of HIF-1αon aberrant cell fate.Usingsingle-cell RNA sequencing(scRNA-seq)and spatial transcriptomics analyses of the HO site,we found that collagen ECM organizationis the most highly up-regulated biological process in MPCs.Zeugopod mesenchymal cell-specific deletion of Hif1α(Hoxa11-CreER^(T2);Hif1a^(fl/fl))significantly mitigated HO in vivo.ScRNA-seq analysis of these Hoxa11-CreER^(T2);Hif1a^(fl/fl)mice identified the PLOD2/LOXpathway for collagen cross-linking as downstream of the HIF-1αregulation of HO.Importantly,our scRNA-seq data and mechanisticstudies further uncovered that glucose metabolism in MPCs is most highly impacted by HIF-1αdeletion.From a translational aspect,a pan-LOX inhibitor significantly decreased HO.A newly screened compound revealed that the inhibition of PLOD2 activity in MPCssignificantly decreased osteogenic differentiation and glycolytic metabolism.This suggests that the HIF-1α/PLOD2/LOX axis linked tometabolism regulates HO-forming MPC fate.These results suggest that the HIF-1α/PLOD2/LOX pathway represents a promisingstrategy to mitigate HO formation.展开更多
Background:Physical activity(PA)is important for cancer survivors.Trials of remotely delivered interventions are needed to assist in reaching under-served non-metropolitan cancer survivors.The objective of this study ...Background:Physical activity(PA)is important for cancer survivors.Trials of remotely delivered interventions are needed to assist in reaching under-served non-metropolitan cancer survivors.The objective of this study was to ascertain whether wearable technology,coupled with health coaching was effective in increasing PA in breast and colorectal cancer survivors living in regional and remote areas in Australia.Methods:Cancer survivors from 5 states were randomized to intervention and control arms.Intervention participants were given a Fitbit Charge 2TMand received up to 6 telephone health coaching sessions.Control participants received PA print materials.Accelerometer assessments at baseline and 12 weeks measured moderate-to-vigorous PA(MVPA),light PA,and sedentary behavior.Results:Eighty-seven participants were recruited(age=63±11 years;74(85%)female).There was a significant net improvement in MVPA of 49.8 min/week,favoring the intervention group(95%confidence interval(95%CI):13.6-86.1,p=0.007).There was also a net increase in MVPA bouts of 39.5 min/week(95%CI:11.9-67.1,p=0.005),favoring the intervention group.Both groups improved light PA and sedentary behavior,but there were no between-group differences.Conclusion:This’s the first study to demonstrate that,when compared to standard practice(i.e.,PA education),a wearable technology intervention coupled with distance-based health coaching,improves MVPA in non-metropolitan cancer survivors.The results display promise for the use of scalable interventions using smart wearable technology in conjunction with phone-based health coaching to foster increased PA in geographically disadvantaged cancer survivors.展开更多
Most modern technologies,such as social media,smart cities,and the internet of things(IoT),rely on big data.When big data is used in the real-world applications,two data challenges such as class overlap and class imba...Most modern technologies,such as social media,smart cities,and the internet of things(IoT),rely on big data.When big data is used in the real-world applications,two data challenges such as class overlap and class imbalance arises.When dealing with large datasets,most traditional classifiers are stuck in the local optimum problem.As a result,it’s necessary to look into new methods for dealing with large data collections.Several solutions have been proposed for overcoming this issue.The rapid growth of the available data threatens to limit the usefulness of many traditional methods.Methods such as oversampling and undersampling have shown great promises in addressing the issues of class imbalance.Among all of these techniques,Synthetic Minority Oversampling TechniquE(SMOTE)has produced the best results by generating synthetic samples for the minority class in creating a balanced dataset.The issue is that their practical applicability is restricted to problems involving tens of thousands or lower instances of each.In this paper,we have proposed a parallel mode method using SMOTE and MapReduce strategy,this distributes the operation of the algorithm among a group of computational nodes for addressing the aforementioned problem.Our proposed solution has been divided into three stages.Thefirst stage involves the process of splitting the data into different blocks using a mapping function,followed by a pre-processing step for each mapping block that employs a hybrid SMOTE algo-rithm for solving the class imbalanced problem.On each map block,a decision tree model would be constructed.Finally,the decision tree blocks would be com-bined for creating a classification model.We have used numerous datasets with up to 4 million instances in our experiments for testing the proposed scheme’s cap-abilities.As a result,the Hybrid SMOTE appears to have good scalability within the framework proposed,and it also cuts down the processing time.展开更多
Halide perovskite light-emitting electrochemical cells are a novel type of the perovskite optoelectronic devices that differs from the perovskite light-emitting diodes by a simple monolayered architecture.Here,we deve...Halide perovskite light-emitting electrochemical cells are a novel type of the perovskite optoelectronic devices that differs from the perovskite light-emitting diodes by a simple monolayered architecture.Here,we develop a perovskite electrochemical cell both for light emission and detection,where the active layer consists of a composite material made of halide perovskite microcrystals,polymer support matrix,and added mobile ions.The perovskite electrochemical cell of CsPbBr3:PEO:LiTFSI composition,emitting light at the wavelength of 523 nm,yields the luminance more than 7000 cd/m2 and electroluminescence efficiency of 4.3 lm/W.The device fabricated on a silicon substrate with transparent single-walled carbon nanotube film as a top contact exhibits 40%lower Joule heating compared to the perovskite optoelectronic devices fabricated on conventional ITO/glass substrates.Moreover,the device operates as a photodetector with a sensitivity up to 0.75 A/W,specific detectivity of 8.56×1011 Jones,and linear dynamic range of 48 dB.The technological potential of such a device is proven by demonstration of 24-pixel indicator display as well as by successful device miniaturization by creation of electroluminescent images with the smallest features less than 50μm.展开更多
Urban-rural integration (URI) is a global challenge that is highly related to inequalities, poverty, economic growth, and other Sustainable Development Goals (SDGs). Existing research has evaluated the extent of URI a...Urban-rural integration (URI) is a global challenge that is highly related to inequalities, poverty, economic growth, and other Sustainable Development Goals (SDGs). Existing research has evaluated the extent of URI and explored its influencing factors, but urban-rural linkages are seldom incorporated in evaluation systems, and geographical factors are rarely recognized as the influencing factors. We construct a URI framework including regional economy, rural development, urban-rural linkage, and urban-rural gap. Based on a dataset consisting of 1,669 counties in China in 2020, we reveal the spatial pattern of URI and find a high correlation between the spatial pattern of URI and the relief degree of land surface (RDLS). Using structural equation modeling, we discover that topography has direct ( − 0.18, p < 0.001) and indirect ( − 0.17, p < 0.001) effects on URI. The indirect negative effects are mediated through the infrastructure, and the combination of localized advantages and modern technical conditions could mitigate the negative impact of topography. Finally, we identify 742 counties as lagging regions in URI, which can be clustered into eight types. Our findings could facilitate policy designing for those countries striving for integrated and sustainable development of urban and rural areas.展开更多
Dear Editor,Suicide amongst the military veteran population is a significant publichealthproblemintheUnitedStates.TheNational VeteranSuicidePreventionAnnualReportrevealedthat6261 died by suicide in 2019[1]. The linger...Dear Editor,Suicide amongst the military veteran population is a significant publichealthproblemintheUnitedStates.TheNational VeteranSuicidePreventionAnnualReportrevealedthat6261 died by suicide in 2019[1]. The lingering effects of the coronavirus disease 2019 (COVID-19) pandemic may account for an increase in veteran suicide rates[1].展开更多
基金funded by the Natural Science Foundation of Fujian Province,China (Grant No.2022J05291)Xiamen Scientific Research Funding for Overseas Chinese Scholars.
文摘Financial time series prediction,whether for classification or regression,has been a heated research topic over the last decade.While traditional machine learning algorithms have experienced mediocre results,deep learning has largely contributed to the elevation of the prediction performance.Currently,the most up-to-date review of advanced machine learning techniques for financial time series prediction is still lacking,making it challenging for finance domain experts and relevant practitioners to determine which model potentially performs better,what techniques and components are involved,and how themodel can be designed and implemented.This review article provides an overview of techniques,components and frameworks for financial time series prediction,with an emphasis on state-of-the-art deep learning models in the literature from2015 to 2023,including standalonemodels like convolutional neural networks(CNN)that are capable of extracting spatial dependencies within data,and long short-term memory(LSTM)that is designed for handling temporal dependencies;and hybrid models integrating CNN,LSTM,attention mechanism(AM)and other techniques.For illustration and comparison purposes,models proposed in recent studies are mapped to relevant elements of a generalized framework comprised of input,output,feature extraction,prediction,and related processes.Among the state-of-the-artmodels,hybrid models like CNNLSTMand CNN-LSTM-AM in general have been reported superior in performance to stand-alone models like the CNN-only model.Some remaining challenges have been discussed,including non-friendliness for finance domain experts,delayed prediction,domain knowledge negligence,lack of standards,and inability of real-time and highfrequency predictions.The principal contributions of this paper are to provide a one-stop guide for both academia and industry to review,compare and summarize technologies and recent advances in this area,to facilitate smooth and informed implementation,and to highlight future research directions.
文摘Correction to:Opto-Electronic Advances https://doi.org/10.29026/oea.2023.220154 published online 26 April 2023 After the publication of this article1,it was brought to our attention that calculations of the PeLEC device elec-troluminescent(EL)efficiency contained a mistake,leading to an inaccurate quantity value.The device’s maxim-um EL efficiency constitutes not‘~120 klm/W’but‘4.3 lm/W’instead.Correction details are listed below.
基金The work was supported by the Ministry of Education and Science of the Russian Federation in the framework of experimental research(Nos.075-01438-22-06 and FSEE-2022-0018)the Russian Science Foundation in theoretical research(No.RSF 23-29-00216).
文摘A new theoretical method to study super-multiperiod superlattices has been developed.The method combines the precision of the 8-band kp-method with the flexibility of the shooting method and the Monte Carlo approach.This method was applied to examine the finest quality samples of super-multiperiod Al_(0.3)Ga_(0.7)As/GaAs superlattices grown by molecular beam epitaxy.The express photoreflectance spectroscopy method was utilized to validate the proposed theoretical method.For the first time,the accurate theoretical analysis of the energy band diagram of super-multiperiod superlattices with experimental verification has been conducted.The proposed approach highly accurately determines transition peak positions and enables the calculation of the energy band diagram,transition energies,relaxation rates,and gain estimation.It has achieved a remarkably low 5%error compared to the commonly used method,which typically results in a 25%error,and allowed to recover the superlattice parameters.The retrieved intrinsic parameters of the samples aligned with XRD data and growth parameters.The proposed method also accurately predicted the escape of the second energy level for quantum well thicknesses less than 5 nm,as was observed in photoreflectance experiments.The new designs of THz light-emitting devices operating at room temperature were suggested by the developed method.
文摘While hypoxic signaling has been shown to play a role in many cellular processes,its role in metabolism-linked extracellular matrix(ECM)organization and downstream processes of cell fate after musculoskeletal injury remains to be determined.Heterotopicossification(HO)is a debilitating condition where abnormal bone formation occurs within extra-skeletal tissues.Hypoxia andhypoxia-inducible factor 1α(HIF-1α)activation have been shown to promote HO.However,the underlying molecular mechanisms bywhich the HIF-1αpathway in mesenchymal progenitor cells(MPCs)contributes to pathologic bone formation remain to beelucidated.Here,we used a proven mouse injury-induced HO model to investigate the role of HIF-1αon aberrant cell fate.Usingsingle-cell RNA sequencing(scRNA-seq)and spatial transcriptomics analyses of the HO site,we found that collagen ECM organizationis the most highly up-regulated biological process in MPCs.Zeugopod mesenchymal cell-specific deletion of Hif1α(Hoxa11-CreER^(T2);Hif1a^(fl/fl))significantly mitigated HO in vivo.ScRNA-seq analysis of these Hoxa11-CreER^(T2);Hif1a^(fl/fl)mice identified the PLOD2/LOXpathway for collagen cross-linking as downstream of the HIF-1αregulation of HO.Importantly,our scRNA-seq data and mechanisticstudies further uncovered that glucose metabolism in MPCs is most highly impacted by HIF-1αdeletion.From a translational aspect,a pan-LOX inhibitor significantly decreased HO.A newly screened compound revealed that the inhibition of PLOD2 activity in MPCssignificantly decreased osteogenic differentiation and glycolytic metabolism.This suggests that the HIF-1α/PLOD2/LOX axis linked tometabolism regulates HO-forming MPC fate.These results suggest that the HIF-1α/PLOD2/LOX pathway represents a promisingstrategy to mitigate HO formation.
基金sponsored by a grant from the Tonkin son Colorectal Cancer Research Fund(#57838)the Ministry of Education,Culture and Sports of Spain for the financing of the Jose Castillejo scholarship(CAS19/00043)to MLR。
文摘Background:Physical activity(PA)is important for cancer survivors.Trials of remotely delivered interventions are needed to assist in reaching under-served non-metropolitan cancer survivors.The objective of this study was to ascertain whether wearable technology,coupled with health coaching was effective in increasing PA in breast and colorectal cancer survivors living in regional and remote areas in Australia.Methods:Cancer survivors from 5 states were randomized to intervention and control arms.Intervention participants were given a Fitbit Charge 2TMand received up to 6 telephone health coaching sessions.Control participants received PA print materials.Accelerometer assessments at baseline and 12 weeks measured moderate-to-vigorous PA(MVPA),light PA,and sedentary behavior.Results:Eighty-seven participants were recruited(age=63±11 years;74(85%)female).There was a significant net improvement in MVPA of 49.8 min/week,favoring the intervention group(95%confidence interval(95%CI):13.6-86.1,p=0.007).There was also a net increase in MVPA bouts of 39.5 min/week(95%CI:11.9-67.1,p=0.005),favoring the intervention group.Both groups improved light PA and sedentary behavior,but there were no between-group differences.Conclusion:This’s the first study to demonstrate that,when compared to standard practice(i.e.,PA education),a wearable technology intervention coupled with distance-based health coaching,improves MVPA in non-metropolitan cancer survivors.The results display promise for the use of scalable interventions using smart wearable technology in conjunction with phone-based health coaching to foster increased PA in geographically disadvantaged cancer survivors.
基金Project(075-15-2022-312) supported by the Ministry of Science and Higher Education of the Russian Federation as part of the World-class Research Center program:Advanced Digital Technologies。
文摘Most modern technologies,such as social media,smart cities,and the internet of things(IoT),rely on big data.When big data is used in the real-world applications,two data challenges such as class overlap and class imbalance arises.When dealing with large datasets,most traditional classifiers are stuck in the local optimum problem.As a result,it’s necessary to look into new methods for dealing with large data collections.Several solutions have been proposed for overcoming this issue.The rapid growth of the available data threatens to limit the usefulness of many traditional methods.Methods such as oversampling and undersampling have shown great promises in addressing the issues of class imbalance.Among all of these techniques,Synthetic Minority Oversampling TechniquE(SMOTE)has produced the best results by generating synthetic samples for the minority class in creating a balanced dataset.The issue is that their practical applicability is restricted to problems involving tens of thousands or lower instances of each.In this paper,we have proposed a parallel mode method using SMOTE and MapReduce strategy,this distributes the operation of the algorithm among a group of computational nodes for addressing the aforementioned problem.Our proposed solution has been divided into three stages.Thefirst stage involves the process of splitting the data into different blocks using a mapping function,followed by a pre-processing step for each mapping block that employs a hybrid SMOTE algo-rithm for solving the class imbalanced problem.On each map block,a decision tree model would be constructed.Finally,the decision tree blocks would be com-bined for creating a classification model.We have used numerous datasets with up to 4 million instances in our experiments for testing the proposed scheme’s cap-abilities.As a result,the Hybrid SMOTE appears to have good scalability within the framework proposed,and it also cuts down the processing time.
基金M.Baeva,A.Vorobyov,V.Neplokh acknowledge the Russian Science Foundation No.22-79-10286(https://rscf.ru/project/22-79-10286/)for supporting silicon substrate processing.D.Gets,APolushkin and S.Makarov acknowledge the Ministry of Science and Higher Education of the Russian Federation(Project 075-15-2021-589)for supporting perovskite synthesisA.G.Nasibulin and D.V.Krasnikov acknowledge the Russian Science Foundation(grant No.20-73-10256)for supporting synthesis of SWCNTs.
文摘Halide perovskite light-emitting electrochemical cells are a novel type of the perovskite optoelectronic devices that differs from the perovskite light-emitting diodes by a simple monolayered architecture.Here,we develop a perovskite electrochemical cell both for light emission and detection,where the active layer consists of a composite material made of halide perovskite microcrystals,polymer support matrix,and added mobile ions.The perovskite electrochemical cell of CsPbBr3:PEO:LiTFSI composition,emitting light at the wavelength of 523 nm,yields the luminance more than 7000 cd/m2 and electroluminescence efficiency of 4.3 lm/W.The device fabricated on a silicon substrate with transparent single-walled carbon nanotube film as a top contact exhibits 40%lower Joule heating compared to the perovskite optoelectronic devices fabricated on conventional ITO/glass substrates.Moreover,the device operates as a photodetector with a sensitivity up to 0.75 A/W,specific detectivity of 8.56×1011 Jones,and linear dynamic range of 48 dB.The technological potential of such a device is proven by demonstration of 24-pixel indicator display as well as by successful device miniaturization by creation of electroluminescent images with the smallest features less than 50μm.
基金the National Natural Science Foundation of China(Grants No.T2261129477 and 41971220)the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDA23070300).
文摘Urban-rural integration (URI) is a global challenge that is highly related to inequalities, poverty, economic growth, and other Sustainable Development Goals (SDGs). Existing research has evaluated the extent of URI and explored its influencing factors, but urban-rural linkages are seldom incorporated in evaluation systems, and geographical factors are rarely recognized as the influencing factors. We construct a URI framework including regional economy, rural development, urban-rural linkage, and urban-rural gap. Based on a dataset consisting of 1,669 counties in China in 2020, we reveal the spatial pattern of URI and find a high correlation between the spatial pattern of URI and the relief degree of land surface (RDLS). Using structural equation modeling, we discover that topography has direct ( − 0.18, p < 0.001) and indirect ( − 0.17, p < 0.001) effects on URI. The indirect negative effects are mediated through the infrastructure, and the combination of localized advantages and modern technical conditions could mitigate the negative impact of topography. Finally, we identify 742 counties as lagging regions in URI, which can be clustered into eight types. Our findings could facilitate policy designing for those countries striving for integrated and sustainable development of urban and rural areas.
文摘Dear Editor,Suicide amongst the military veteran population is a significant publichealthproblemintheUnitedStates.TheNational VeteranSuicidePreventionAnnualReportrevealedthat6261 died by suicide in 2019[1]. The lingering effects of the coronavirus disease 2019 (COVID-19) pandemic may account for an increase in veteran suicide rates[1].
