In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performan...In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performance of recommendation systems.Due to the imbalance,machine learning algorithms tend to classify inputs into the positive(majority)class every time to achieve high prediction accuracy.Imbalance can be categorized such as by features and classes,but most studies consider only class imbalance.In this paper,we propose a recommendation system that can integrate multiple networks to adapt to a large number of imbalanced features and can deal with highly skewed and imbalanced datasets through a loss function.We propose a loss aware feature attention mechanism(LAFAM)to solve the issue of feature imbalance.The network incorporates an attention mechanism and uses multiple sub-networks to classify and learn features.For better results,the network can learn the weights of sub-networks and assign higher weights to important features.We propose suppression loss to address class imbalance,which favors negative loss by penalizing positive loss,and pays more attention to sample points near the decision boundary.Experiments on two large-scale datasets verify that the performance of the proposed system is greatly improved compared to baseline methods.展开更多
Climate warming causes mountainous species to shift their distributions towards higher elevations.How elevation influences growth-climate relationship in mountain regions has been intensively investigated.However,how ...Climate warming causes mountainous species to shift their distributions towards higher elevations.How elevation influences growth-climate relationship in mountain regions has been intensively investigated.However,how microtopography shapes tree growth and its drought resistance along the elevation gradient remains poorly understood.We used a network of Larix principis-rupprechtii tree-ring data comprising 1,918 trees from different age classes and mountain slopes,along an elevation gradient ranging from 970 to 1,869 m,to investigate how slope gradients mediate the growth and drought resilience of larch trees along an elevation gradient in North China.Growing season drought and temperature were the major limiting climatic factors for larch trees across the study region.Larch trees younger than 40 years exhibited a stronger positive correlation between basal area increment(BAI)and elevation on steep slopes(10°-35°)than on flat(0°-5°)or gentle(5°-10°)slopes.At low-elevation steep slopes,the growth of larch trees younger than 40 years showed a stronger correlation with the Palmer drought severity index(PDSI).Both resistance and resilience were found to increase along the elevation gradient on steep slopes for young larch trees but not for old larch trees.No significant differences were observed in the drought recovery ability of larch trees across all age groups at increasing elevation.Our results highlight that drought events may particularly affect the growth of young larch trees on low-elevation steep slopes,with potential repercussions on mortality rates.展开更多
The magnetization dynamics of lanthanide coordination compounds are fundamentals governing their potential applications such as information storage or molecular switches.Herein,two two-dimensional coordination polymer...The magnetization dynamics of lanthanide coordination compounds are fundamentals governing their potential applications such as information storage or molecular switches.Herein,two two-dimensional coordination polymers[Er(CA)_(1.5)(bpy)(DMF)]_(n)(1)and[Er(CA)_(1.5)(phen)(DMF)]_(n)(2)(H_(2)CA=2,5-dichloro-3,6-dihydroxy-p-quinone,bpy=2,2'-bipyridine,phen=1,10-phenanthroline)were synthesized and fully characterized.By the irradiation of ultraviolet light,1 and 2 were converted to la and 2a which contain light-generated radicals,inducing an increase ofχ_(MT)at room temperature.A detailed study of the dynamic magnetic property shows that the magnetization dynamics observed for 1 and la are dominated by Raman process,but Orbach and Raman processes are observed in 2 and 2a.The structural factors influencing the magnetic properties of this photomagnetic system are discussed.展开更多
Aspect-Based Sentiment Analysis(ABSA)is a fundamental area of research in Natural Language Processing(NLP).Within ABSA,Aspect Sentiment Quad Prediction(ASQP)aims to accurately identify sentiment quadruplets in target ...Aspect-Based Sentiment Analysis(ABSA)is a fundamental area of research in Natural Language Processing(NLP).Within ABSA,Aspect Sentiment Quad Prediction(ASQP)aims to accurately identify sentiment quadruplets in target sentences,including aspect terms,aspect categories,corresponding opinion terms,and sentiment polarity.However,most existing research has focused on English datasets.Consequently,while ASQP has seen significant progress in English,the Chinese ASQP task has remained relatively stagnant.Drawing inspiration from methods applied to English ASQP,we propose Chinese generation templates and employ prompt-based instruction learning to enhance the model’s understanding of the task,ultimately improving ASQP performance in the Chinese context.Ultimately,under the same pre-training model configuration,our approach achieved a 5.79%improvement in the F1 score compared to the previously leading method.Furthermore,when utilizing a larger model with reduced training parameters,the F1 score demonstrated an 8.14%enhancement.Additionally,we suggest a novel evaluation metric based on the characteristics of generative models,better-reflecting model generalization.Experimental results validate the effectiveness of our approach.展开更多
Light-induced transformation from diamagnetic ligand to paramagnetic radical offers a unique approach to modulating the magnetization dynamics of magnetic compounds.In this study,we present a two-dimensional Yb(III)co...Light-induced transformation from diamagnetic ligand to paramagnetic radical offers a unique approach to modulating the magnetization dynamics of magnetic compounds.In this study,we present a two-dimensional Yb(III)coordination polymer 1 and its Y(III)-diluted analog 1@Y,along with their lightinduced radical forms,1a and 1a@Y.Due to the presence of light-induced radicals,the magnetic susceptibility properties(_(χM)T values)of 1a and 1a@Y featured remarkable 82.3%and 88.9%enhancements,respectively,at room temperature,compared with those of 1 and 1@Y,and were accompanied by an unexpected increase in the effective energy barrier for magnetization dynamics.The magnetization dynamics of 1 and 1a involved both Orbach and Raman processes,while 1@Y and 1a@Y exhibited two Raman processes,indicating that the Orbach process that occurred in 1 and 1a was mainly attributed to the magnetic coupling between Yb(III)ions.This work showcases the potential of light-induced radical transformation for tuning both the static and dynamic magnetic properties of lanthanide coordination polymers.展开更多
Perfluorooctanoic acid(PFOA),a persistent environmental pollutant,poses serious health risks and remains challenging to detect due to its low light absorption and weak binding affinity.Luminescence sensing offers a po...Perfluorooctanoic acid(PFOA),a persistent environmental pollutant,poses serious health risks and remains challenging to detect due to its low light absorption and weak binding affinity.Luminescence sensing offers a powerful approach to its rapid and quantitative detection,with electron transfer playing a critical role when energy transfer is limited by the analyte’s unfavorable energy-level alignment.Lanthanidebased metal–organic frameworks(MOFs),featuring sharp emissions,long-lived excited states,and tunable electronic structures,are particularly promising for such applications owing to their ability to precisely regulate electron transfer pathways through structural modulation.However,the impact of the coordination mode and symmetry on sensing performance remains poorly understood due to its complex mechanisms.Herein,we report a series of eight europiumbased MOFs,Eu_(2)(FDA)_(3)(solvent)x,synthesized using a single ligand to enable systematic investigation of the relationship between Eu^(3+) coordination symmetry and PFOA sensing behavior.Notably,MOFs with lower coordination symmetry exhibit significantly higher quenching efficiency due to increased energy-level splitting that promotes electron transfer.This study establishes a clear structurefunction correlation and provides valuable insights for the rational design of MOF-based sensors targeting spectroscopically challenging contaminants,thereby deepening the understanding of luminescence sensing mechanisms based on structure-driven signal transduction processes.展开更多
The electron-donating unit 2,3,4,6-tetrahydro-1,6-dithia-3a-azaphenalene (THDTAP) was introduced onto terpyridine (TPy) to give a donor-acceptor (D-A) type TPy-ligand (compound 2).Upon selective oxidation of two sulfu...The electron-donating unit 2,3,4,6-tetrahydro-1,6-dithia-3a-azaphenalene (THDTAP) was introduced onto terpyridine (TPy) to give a donor-acceptor (D-A) type TPy-ligand (compound 2).Upon selective oxidation of two sulfur atoms on the THDTAP moiety of 2,the ligands 3-6 were created.The electronic structures of 2-6 were evaluated by theoretical,electrochemical,and spectroscopic investigations.The oxidation on the sulfur atoms brings significant influence on the electron-donating ability of THDTAP moiety,subsequently,leads to fine modulations on intramolecular charge-transfer (ICT) of 2-6 and the electronically excited states of the complexes of 2-6 with metal ions.Based on the optical response of 2-6 toward metal ions,the step-by-step recognition of Zn2+,Cd2+,and Ag+ ions is set up by employing 2 and 3 as combined fluorescence sensors.展开更多
基金supported by the National Key Research and Development Program of China(Grant numbers:2021YFF0901705,2021YFF0901700)the State Key Laboratory of Media Convergence and Communication,Communication University of China+1 种基金the Fundamental Research Funds for the Central Universitiesthe High-Quality and Cutting-Edge Disciplines Construction Project for Universities in Beijing(Internet Information,Communication University of China).
文摘In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performance of recommendation systems.Due to the imbalance,machine learning algorithms tend to classify inputs into the positive(majority)class every time to achieve high prediction accuracy.Imbalance can be categorized such as by features and classes,but most studies consider only class imbalance.In this paper,we propose a recommendation system that can integrate multiple networks to adapt to a large number of imbalanced features and can deal with highly skewed and imbalanced datasets through a loss function.We propose a loss aware feature attention mechanism(LAFAM)to solve the issue of feature imbalance.The network incorporates an attention mechanism and uses multiple sub-networks to classify and learn features.For better results,the network can learn the weights of sub-networks and assign higher weights to important features.We propose suppression loss to address class imbalance,which favors negative loss by penalizing positive loss,and pays more attention to sample points near the decision boundary.Experiments on two large-scale datasets verify that the performance of the proposed system is greatly improved compared to baseline methods.
基金funded by the National Natural Science Foundation of China(No.U24A20353)the S&T Program of Hebei(Nos.226Z6801G,C2021204002,and 20210365)+1 种基金the Talent Introduction Program in Hebei Agricultural University(No.YJ201918)supported by the SERI-funded ERC Starting Grant,project MB23.00011.
文摘Climate warming causes mountainous species to shift their distributions towards higher elevations.How elevation influences growth-climate relationship in mountain regions has been intensively investigated.However,how microtopography shapes tree growth and its drought resistance along the elevation gradient remains poorly understood.We used a network of Larix principis-rupprechtii tree-ring data comprising 1,918 trees from different age classes and mountain slopes,along an elevation gradient ranging from 970 to 1,869 m,to investigate how slope gradients mediate the growth and drought resilience of larch trees along an elevation gradient in North China.Growing season drought and temperature were the major limiting climatic factors for larch trees across the study region.Larch trees younger than 40 years exhibited a stronger positive correlation between basal area increment(BAI)and elevation on steep slopes(10°-35°)than on flat(0°-5°)or gentle(5°-10°)slopes.At low-elevation steep slopes,the growth of larch trees younger than 40 years showed a stronger correlation with the Palmer drought severity index(PDSI).Both resistance and resilience were found to increase along the elevation gradient on steep slopes for young larch trees but not for old larch trees.No significant differences were observed in the drought recovery ability of larch trees across all age groups at increasing elevation.Our results highlight that drought events may particularly affect the growth of young larch trees on low-elevation steep slopes,with potential repercussions on mortality rates.
基金Project supported by the National Natural Science Foundation of China(21971123,21931004,92156002)。
文摘The magnetization dynamics of lanthanide coordination compounds are fundamentals governing their potential applications such as information storage or molecular switches.Herein,two two-dimensional coordination polymers[Er(CA)_(1.5)(bpy)(DMF)]_(n)(1)and[Er(CA)_(1.5)(phen)(DMF)]_(n)(2)(H_(2)CA=2,5-dichloro-3,6-dihydroxy-p-quinone,bpy=2,2'-bipyridine,phen=1,10-phenanthroline)were synthesized and fully characterized.By the irradiation of ultraviolet light,1 and 2 were converted to la and 2a which contain light-generated radicals,inducing an increase ofχ_(MT)at room temperature.A detailed study of the dynamic magnetic property shows that the magnetization dynamics observed for 1 and la are dominated by Raman process,but Orbach and Raman processes are observed in 2 and 2a.The structural factors influencing the magnetic properties of this photomagnetic system are discussed.
基金supported by the National Key Research and Development Program(Nos.2021YFF0901705,2021YFF0901700)the State Key Laboratory of Media Convergence and Communication,Communication University of China+1 种基金the Fundamental Research Funds for the Central Universitiesthe High-Quality and Cutting-Edge Disciplines Construction Project for Universities in Beijing(Internet Information,Communication University of China).
文摘Aspect-Based Sentiment Analysis(ABSA)is a fundamental area of research in Natural Language Processing(NLP).Within ABSA,Aspect Sentiment Quad Prediction(ASQP)aims to accurately identify sentiment quadruplets in target sentences,including aspect terms,aspect categories,corresponding opinion terms,and sentiment polarity.However,most existing research has focused on English datasets.Consequently,while ASQP has seen significant progress in English,the Chinese ASQP task has remained relatively stagnant.Drawing inspiration from methods applied to English ASQP,we propose Chinese generation templates and employ prompt-based instruction learning to enhance the model’s understanding of the task,ultimately improving ASQP performance in the Chinese context.Ultimately,under the same pre-training model configuration,our approach achieved a 5.79%improvement in the F1 score compared to the previously leading method.Furthermore,when utilizing a larger model with reduced training parameters,the F1 score demonstrated an 8.14%enhancement.Additionally,we suggest a novel evaluation metric based on the characteristics of generative models,better-reflecting model generalization.Experimental results validate the effectiveness of our approach.
基金supported by the National Natural Science Foundation of China(grant nos.22435002a,92156002,21971123,and 62401563)L.U.acknowledges the financial support of the research projects A-800079-00-00,A-8000017-00-00,and A-8001894-00-00 of the National University of Singapore(NUS)+1 种基金Ab initio calculations were done on the Advanced Supercomputer for Petascale Innovation,Research and Enterprise 2A(ASPIRE-2A)cluster of the Singapore National Supercomputing Centre(NSCCwww.nscc.sg)under project 11001278.The highperformance computational(HPC)resources of NUS(HPC-NUS)are gratefully acknowledged.The NUS Research Scholarship program supported W.H.
文摘Light-induced transformation from diamagnetic ligand to paramagnetic radical offers a unique approach to modulating the magnetization dynamics of magnetic compounds.In this study,we present a two-dimensional Yb(III)coordination polymer 1 and its Y(III)-diluted analog 1@Y,along with their lightinduced radical forms,1a and 1a@Y.Due to the presence of light-induced radicals,the magnetic susceptibility properties(_(χM)T values)of 1a and 1a@Y featured remarkable 82.3%and 88.9%enhancements,respectively,at room temperature,compared with those of 1 and 1@Y,and were accompanied by an unexpected increase in the effective energy barrier for magnetization dynamics.The magnetization dynamics of 1 and 1a involved both Orbach and Raman processes,while 1@Y and 1a@Y exhibited two Raman processes,indicating that the Orbach process that occurred in 1 and 1a was mainly attributed to the magnetic coupling between Yb(III)ions.This work showcases the potential of light-induced radical transformation for tuning both the static and dynamic magnetic properties of lanthanide coordination polymers.
基金supported by the Robert A.Welch Foundation through an endowed chair(grant no.A-0030)to H.-C.Z.the National Natural Science Foundation of China(grant nos.22471130,22261132509,22435002,and 22121005).
文摘Perfluorooctanoic acid(PFOA),a persistent environmental pollutant,poses serious health risks and remains challenging to detect due to its low light absorption and weak binding affinity.Luminescence sensing offers a powerful approach to its rapid and quantitative detection,with electron transfer playing a critical role when energy transfer is limited by the analyte’s unfavorable energy-level alignment.Lanthanidebased metal–organic frameworks(MOFs),featuring sharp emissions,long-lived excited states,and tunable electronic structures,are particularly promising for such applications owing to their ability to precisely regulate electron transfer pathways through structural modulation.However,the impact of the coordination mode and symmetry on sensing performance remains poorly understood due to its complex mechanisms.Herein,we report a series of eight europiumbased MOFs,Eu_(2)(FDA)_(3)(solvent)x,synthesized using a single ligand to enable systematic investigation of the relationship between Eu^(3+) coordination symmetry and PFOA sensing behavior.Notably,MOFs with lower coordination symmetry exhibit significantly higher quenching efficiency due to increased energy-level splitting that promotes electron transfer.This study establishes a clear structurefunction correlation and provides valuable insights for the rational design of MOF-based sensors targeting spectroscopically challenging contaminants,thereby deepening the understanding of luminescence sensing mechanisms based on structure-driven signal transduction processes.
基金The authors ack no wledge the grant from the Nati onal Natural Science Foundation of China (Nos. 21871119, 21522203)the National Key R&D Program of China (No. 2017YFA0204903).
文摘The electron-donating unit 2,3,4,6-tetrahydro-1,6-dithia-3a-azaphenalene (THDTAP) was introduced onto terpyridine (TPy) to give a donor-acceptor (D-A) type TPy-ligand (compound 2).Upon selective oxidation of two sulfur atoms on the THDTAP moiety of 2,the ligands 3-6 were created.The electronic structures of 2-6 were evaluated by theoretical,electrochemical,and spectroscopic investigations.The oxidation on the sulfur atoms brings significant influence on the electron-donating ability of THDTAP moiety,subsequently,leads to fine modulations on intramolecular charge-transfer (ICT) of 2-6 and the electronically excited states of the complexes of 2-6 with metal ions.Based on the optical response of 2-6 toward metal ions,the step-by-step recognition of Zn2+,Cd2+,and Ag+ ions is set up by employing 2 and 3 as combined fluorescence sensors.
