Nowadays,ensuring thequality of networkserviceshas become increasingly vital.Experts are turning toknowledge graph technology,with a significant emphasis on entity extraction in the identification of device configurat...Nowadays,ensuring thequality of networkserviceshas become increasingly vital.Experts are turning toknowledge graph technology,with a significant emphasis on entity extraction in the identification of device configurations.This research paper presents a novel entity extraction method that leverages a combination of active learning and attention mechanisms.Initially,an improved active learning approach is employed to select the most valuable unlabeled samples,which are subsequently submitted for expert labeling.This approach successfully addresses the problems of isolated points and sample redundancy within the network configuration sample set.Then the labeled samples are utilized to train the model for network configuration entity extraction.Furthermore,the multi-head self-attention of the transformer model is enhanced by introducing the Adaptive Weighting method based on the Laplace mixture distribution.This enhancement enables the transformer model to dynamically adapt its focus to words in various positions,displaying exceptional adaptability to abnormal data and further elevating the accuracy of the proposed model.Through comparisons with Random Sampling(RANDOM),Maximum Normalized Log-Probability(MNLP),Least Confidence(LC),Token Entrop(TE),and Entropy Query by Bagging(EQB),the proposed method,Entropy Query by Bagging and Maximum Influence Active Learning(EQBMIAL),achieves comparable performance with only 40% of the samples on both datasets,while other algorithms require 50% of the samples.Furthermore,the entity extraction algorithm with the Adaptive Weighted Multi-head Attention mechanism(AW-MHA)is compared with BILSTM-CRF,Mutil_Attention-Bilstm-Crf,Deep_Neural_Model_NER and BERT_Transformer,achieving precision rates of 75.98% and 98.32% on the two datasets,respectively.Statistical tests demonstrate the statistical significance and effectiveness of the proposed algorithms in this paper.展开更多
Named entity recognition(NER)is a fundamental task of information extraction(IE),and it has attracted considerable research attention in recent years.The abundant annotated English NER datasets have significantly prom...Named entity recognition(NER)is a fundamental task of information extraction(IE),and it has attracted considerable research attention in recent years.The abundant annotated English NER datasets have significantly promoted the NER research in the English field.By contrast,much fewer efforts are made to the Chinese NER research,especially in the scientific domain,due to the scarcity of Chinese NER datasets.To alleviate this problem,we present aChinese scientificNER dataset–SciCN,which contains entity annotations of titles and abstracts derived from 3,500 scientific papers.We manually annotate a total of 62,059 entities,and these entities are classified into six types.Compared to English scientific NER datasets,SciCN has a larger scale and is more diverse,for it not only contains more paper abstracts but these abstracts are derived from more research fields.To investigate the properties of SciCN and provide baselines for future research,we adapt a number of previous state-of-theart Chinese NER models to evaluate SciCN.Experimental results show that SciCN is more challenging than other Chinese NER datasets.In addition,previous studies have proven the effectiveness of using lexicons to enhance Chinese NER models.Motivated by this fact,we provide a scientific domain-specific lexicon.Validation results demonstrate that our lexicon delivers better performance gains than lexicons of other domains.We hope that the SciCN dataset and the lexicon will enable us to benchmark the NER task regarding the Chinese scientific domain and make progress for future research.The dataset and lexicon are available at:https://github.com/yangjingla/SciCN.git.展开更多
Named Entity Recognition(NER)stands as a fundamental task within the field of biomedical text mining,aiming to extract specific types of entities such as genes,proteins,and diseases from complex biomedical texts and c...Named Entity Recognition(NER)stands as a fundamental task within the field of biomedical text mining,aiming to extract specific types of entities such as genes,proteins,and diseases from complex biomedical texts and categorize them into predefined entity types.This process can provide basic support for the automatic construction of knowledge bases.In contrast to general texts,biomedical texts frequently contain numerous nested entities and local dependencies among these entities,presenting significant challenges to prevailing NER models.To address these issues,we propose a novel Chinese nested biomedical NER model based on RoBERTa and Global Pointer(RoBGP).Our model initially utilizes the RoBERTa-wwm-ext-large pretrained language model to dynamically generate word-level initial vectors.It then incorporates a Bidirectional Long Short-Term Memory network for capturing bidirectional semantic information,effectively addressing the issue of long-distance dependencies.Furthermore,the Global Pointer model is employed to comprehensively recognize all nested entities in the text.We conduct extensive experiments on the Chinese medical dataset CMeEE and the results demonstrate the superior performance of RoBGP over several baseline models.This research confirms the effectiveness of RoBGP in Chinese biomedical NER,providing reliable technical support for biomedical information extraction and knowledge base construction.展开更多
Chinese named entity recognition(CNER)has received widespread attention as an important task of Chinese information extraction.Most previous research has focused on individually studying flat CNER,overlapped CNER,or d...Chinese named entity recognition(CNER)has received widespread attention as an important task of Chinese information extraction.Most previous research has focused on individually studying flat CNER,overlapped CNER,or discontinuous CNER.However,a unified CNER is often needed in real-world scenarios.Recent studies have shown that grid tagging-based methods based on character-pair relationship classification hold great potential for achieving unified NER.Nevertheless,how to enrich Chinese character-pair grid representations and capture deeper dependencies between character pairs to improve entity recognition performance remains an unresolved challenge.In this study,we enhance the character-pair grid representation by incorporating both local and global information.Significantly,we introduce a new approach by considering the character-pair grid representation matrix as a specialized image,converting the classification of character-pair relationships into a pixel-level semantic segmentation task.We devise a U-shaped network to extract multi-scale and deeper semantic information from the grid image,allowing for a more comprehensive understanding of associative features between character pairs.This approach leads to improved accuracy in predicting their relationships,ultimately enhancing entity recognition performance.We conducted experiments on two public CNER datasets in the biomedical domain,namely CMeEE-V2 and Diakg.The results demonstrate the effectiveness of our approach,which achieves F1-score improvements of 7.29 percentage points and 1.64 percentage points compared to the current state-of-the-art(SOTA)models,respectively.展开更多
In recent years,cyber attacks have been intensifying and causing great harm to individuals,companies,and countries.The mining of cyber threat intelligence(CTI)can facilitate intelligence integration and serve well in ...In recent years,cyber attacks have been intensifying and causing great harm to individuals,companies,and countries.The mining of cyber threat intelligence(CTI)can facilitate intelligence integration and serve well in combating cyber attacks.Named Entity Recognition(NER),as a crucial component of text mining,can structure complex CTI text and aid cybersecurity professionals in effectively countering threats.However,current CTI NER research has mainly focused on studying English CTI.In the limited studies conducted on Chinese text,existing models have shown poor performance.To fully utilize the power of Chinese pre-trained language models(PLMs)and conquer the problem of lengthy infrequent English words mixing in the Chinese CTIs,we propose a residual dilated convolutional neural network(RDCNN)with a conditional random field(CRF)based on a robustly optimized bidirectional encoder representation from transformers pre-training approach with whole word masking(RoBERTa-wwm),abbreviated as RoBERTa-wwm-RDCNN-CRF.We are the first to experiment on the relevant open source dataset and achieve an F1-score of 82.35%,which exceeds the common baseline model bidirectional encoder representation from transformers(BERT)-bidirectional long short-term memory(BiLSTM)-CRF in this field by about 19.52%and exceeds the current state-of-the-art model,BERT-RDCNN-CRF,by about 3.53%.In addition,we conducted an ablation study on the encoder part of the model to verify the effectiveness of the proposed model and an in-depth investigation of the PLMs and encoder part of the model to verify the effectiveness of the proposed model.The RoBERTa-wwm-RDCNN-CRF model,the shared pre-processing,and augmentation methods can serve the subsequent fundamental tasks such as cybersecurity information extraction and knowledge graph construction,contributing to important applications in downstream tasks such as intrusion detection and advanced persistent threat(APT)attack detection.展开更多
由于电离层电子密度随时间变化,且空间分布不均匀,对不同频段的无线电波产生延缓和折射,因此电离层电子密度变化是影响短波通信、卫星通信、全球导航卫星系统和其他空间通信质量的一个主要因素,本文对全球电离层电子密度(Number of elec...由于电离层电子密度随时间变化,且空间分布不均匀,对不同频段的无线电波产生延缓和折射,因此电离层电子密度变化是影响短波通信、卫星通信、全球导航卫星系统和其他空间通信质量的一个主要因素,本文对全球电离层电子密度(Number of electron,Ne)的预测工作对短波通信设备三维射线实时追踪定位提供必要条件。本文采用国际电离层参考模型提供的2016年电离层Ne数据,根据数据的三维空间时间序列特征,搭建了自编码器和卷积长短期记忆(Convolutional Long Short-Term Memory Network,Conv LSTM)网络组成的网络结构,在不引入地球自转周期之外任何先验知识的条件下,对Ne数据进行深度学习并实现预测,首先通过实验对比了SGD、Adagrad、Adadelta、Adam、Adamax和Nadam六种优化算法的性能,又对比了三种预测策略的均方根误差(Root Mean Square Error, RMSE),1h-to-1h预测策略的全球平均RMSE为1.0 NEU(最大值的0.4%),1h-to-24h和24h-to-24h预测策略的全球平均RMSE为6.3 NEU(2.6%)。由实验结果得出以下结论,一是Nadam优化算法更适合电离层Ne的深度学习,二是1h预测策略的性能与之前类似的电离层TEC预测工作(RMSE高于1.5 TECU,最大值的1%)相比有竞争力,但预测时间太短且对数据的实时性要求较高,三是两种24h预测策略虽能实现长期预测但性能不理想,要实现三维空间时间序列的长期高精度预测需要进一步改善神经网络、模型结构和预测策略。展开更多
With the rapid development of information technology,the electronifi-cation of medical records has gradually become a trend.In China,the population base is huge and the supporting medical institutions are numerous,so ...With the rapid development of information technology,the electronifi-cation of medical records has gradually become a trend.In China,the population base is huge and the supporting medical institutions are numerous,so this reality drives the conversion of paper medical records to electronic medical records.Electronic medical records are the basis for establishing a smart hospital and an important guarantee for achieving medical intelligence,and the massive amount of electronic medical record data is also an important data set for conducting research in the medical field.However,electronic medical records contain a large amount of private patient information,which must be desensitized before they are used as open resources.Therefore,to solve the above problems,data masking for Chinese electronic medical records with named entity recognition is proposed in this paper.Firstly,the text is vectorized to satisfy the required format of the model input.Secondly,since the input sentences may have a long or short length and the relationship between sentences in context is not negligible.To this end,a neural network model for named entity recognition based on bidirectional long short-term memory(BiLSTM)with conditional random fields(CRF)is constructed.Finally,the data masking operation is performed based on the named entity recog-nition results,mainly using regular expression filtering encryption and principal component analysis(PCA)word vector compression and replacement.In addi-tion,comparison experiments with the hidden markov model(HMM)model,LSTM-CRF model,and BiLSTM model are conducted in this paper.The experi-mental results show that the method used in this paper achieves 92.72%Accuracy,92.30%Recall,and 92.51%F1_score,which has higher accuracy compared with other models.展开更多
With its complex nonlinear dynamic behavior,the tristable system has shown excellent performance in areas such as energy harvesting and vibration suppression,and has attracted a lot of attention.In this paper,an asymm...With its complex nonlinear dynamic behavior,the tristable system has shown excellent performance in areas such as energy harvesting and vibration suppression,and has attracted a lot of attention.In this paper,an asymmetric tristable design is proposed to improve the vibration suppression efficiency of nonlinear energy sinks(NESs)for the first time.The proposed asymmetric tristable NES(ATNES)is composed of a pair of oblique springs and a vertical spring.Then,the three stable states,symmetric and asymmetric,can be achieved by the adjustment of the distance and stiffness asymmetry of the oblique springs.The governing equations of a linear oscillator(LO)coupled with the ATNES are derived.The approximate analytical solution to the coupled system is obtained by the harmonic balance method(HBM)and verified numerically.The vibration suppression efficiency of three types of ATNES is compared.The results show that the asymmetric design can improve the efficiency of vibration reduction through comparing the chaotic motion of the NES oscillator between asymmetric steady states.In addition,compared with the symmetrical tristable NES(TNES),the ATNES can effectively control smaller structural vibrations.In other words,the ATNES can effectively solve the threshold problem of TNES failure to weak excitation.Therefore,this paper reveals the vibration reduction mechanism of the ATNES,and provides a pathway to expand the effective excitation amplitude range of the NES.展开更多
The extraction and understanding of text knowledge become increasingly crucial in the age of big data.One of the current research areas in the field of natural language processing(NLP)is how to accurately understand t...The extraction and understanding of text knowledge become increasingly crucial in the age of big data.One of the current research areas in the field of natural language processing(NLP)is how to accurately understand the text and collect accurate linguistic information because Chinese vocabulary is diverse and ambiguous.This paper mainly studies the candidate entity generation module of the entity link system.The candidate entity generation module constructs an entity reference expansion algorithm to improve the recall rate of candidate entities.In order to improve the efficiency of the connection algorithm of the entire system while ensuring the recall rate of candidate entities,we design a graph model filtering algorithm that fuses shallow semantic information to filter the list of candidate entities,and verify and analyze the efficiency of the algorithm through experiments.By analyzing the related technology of the entity linking algorithm,we study the related technology of candidate entity generation and entity disambiguation,improve the traditional entity linking algorithm,and give an innovative and practical entity linking model.The recall rate exceeds 82%,and the link accuracy rate exceeds 73%.Efficient and accurate entity linking can help machines to better understand text semantics,further promoting the development of NLP and improving the users’knowledge acquisition experience on the text.展开更多
The natural and economic conditions of agricultural resources vary greatly in China,especially in hilly and mountainous areas.The phenomenon of land fragmentation has become increasingly prominent,so that large-scale ...The natural and economic conditions of agricultural resources vary greatly in China,especially in hilly and mountainous areas.The phenomenon of land fragmentation has become increasingly prominent,so that large-scale operations cannot be fully achieved in a short period of time,and the centralized and continuous scale of operations cannot be realized in China.In addition,with increasingly strict farmland protection and agricultural land use control systems,the issues of nongrain production and nonagricultural cultivated land use have become increasingly restricted.Thus,it is quite necessary to determine the appropriate scale of operations and the means to achieve moderately scaled operations for the new business entities.On the basis of microsurvey data for 108 new citrus-planting business entities in the modern agricultural park in the Chongqing's Jiangjin District,an area with long citrus-planting history,we measured the moderate scale of new citrus-planting business entities to maximize profit using a translog growth model.According to the projection pursuit model,we evaluated the suitability conditions of citrus planting in garden land,woodland,grassland,and general cultivated land in the study area.We then explored the potential for new moderate-scale business entities within different farming radii.The results showed that large-scale planting conditions of citrus in garden land,forest land,grassland,and general cultivated land in the study area were suitable,and the proportion of high-suitable and mediumsuitable land was 73.42%.Under the existing social and economic conditions,the moderate scale of new citrus-planting business entities in the study area was 1.8–2.7 hm^(2).In particular,its rankings from large to small were agricultural enterprises(17.19–25.78 hm^(2)),farmer cooperatives(16.88–25.33 hm^(2)),big growers and breeders(6.39–9.59 hm^(2)),and family farms(5.02–7.53 hm^(2)).In the sample of 108 households,only 47.22%of the entities achieved moderate-scale operation,of which 25%of the entities achieved a scale greater than moderate operation.However,52.78%of the entities achieved a scale of less than moderate operation.These entities would have to transfer the surrounding adjacent garden land,forest land,grassland,or general cultivated land to achieve largescale land management.The land area of the adjacent gardens in a 1000 m farming radius could meet the moderate-scale operation demand of the vast majority of new business entities.For the vast majority of new business entities,the land area of the adjacent woodland,grassland,and general cultivated land could supplement the garden land to achieve moderate-scale operation in a 500 m farming radius.If the land area with moderate suitability and high suitability is prioritized,the land area in the adjacent gardens in the 1000 m farming radius could meet the moderate-scale operation demand for the vast majority of new business entities.Within the 500 m farming radius,the vast majority of new business entities have achieved moderate-scale operations if the land area adjacent to forest land,grassland,and general arable land was supplemented by gardens;however,a few entities could not achieve moderate-scale operations.From the village perspective,gardens were prioritized.Sixteen villages had planting areas and planting suitability that exceeded the average level of the study area,accounting for 23.53%.If combined with the reserve potential of the garden land,eight villages could improve their potential,accounting for 11.76%.Therefore,the conditions of large-scale citrus planting in the study area should be further improved,and the scale expansion of new citrus-planting business entities should receive additional scientific guidance.展开更多
The positive impact of the development of new agricultural business entities and their institutional systems on the optimization of rural governance structures can be examined from various perspectives,including the s...The positive impact of the development of new agricultural business entities and their institutional systems on the optimization of rural governance structures can be examined from various perspectives,including the state,market,rural society,urban-rural relations,and grassroots governance structure in rural communities.The development of these new agricultural business entities has not only redefined but also restructured the power distribution and governance patterns in rural developing countries,markets,and villages.The rural social order has evolved into a ternary mutual structure governance pattern,often referred to as the“state market rural”model.This transformation has prompted adjustments in the national economic and social policy structure and management systems at both macro and micro levels.It has led to the reshaping of power dynamics,benefit distribution,and governance structures in both urban and rural areas,resulting in significant changes to the economic and social fabric of rural regions.Furthermore,the grassroots governance structure in rural society,characterized by“township governance and village governance,”is undergoing continuous development and improvement.This transition is marked by a shift towards a collaborative governance structure that encourages diverse participation.Building upon the aforementioned optimizations,the rural governance structure now exhibits new characteristics.These include a more extensive and diverse range of rural governance mechanisms,increased openness in governance processes,and a heightened synergy among various governance mechanisms.This dynamic evolution reflects a richer,more diverse,and more open approach to rural governance,fostering a stronger collaborative effort in the pursuit of effective governance.展开更多
The post-collisional Cenozoic basic volcanic rocks in NE Turkey show temporal variations in whole-rock lithophile element and highly siderophile element(HSE)systematics that are mainly associated with the nature of su...The post-collisional Cenozoic basic volcanic rocks in NE Turkey show temporal variations in whole-rock lithophile element and highly siderophile element(HSE)systematics that are mainly associated with the nature of sub-continental lithospheric mantle(SCLM)sources and parental melt generation.So far,the traditional whole-rock lithophile geochemical data of these basic volcanic rocks have provided important constraints on the nature of SCLM sources.Integrated lithophile element and HSE geochemical data of these basic volcanic rocks also reveal the heterogeneity of the SCLM source,which is principally related to variable metasomatism resulting from previous subduction(s)and post-collisional mantle-crust interactions in an extensional setting.Lithophile element geochemical features suggest that the parental magmas have derived from metasomatized spinel-to garnet-bearing SCLM sources for Eocene and Miocene basic volcanic rocks with subduction signatures whereas originated from spinel-to garnet-bearing SCLM sources for Mio-Pliocene and Plio-Quaternary basaltic volcanic rocks without the subduction signature.Lithophile element and HSE geo-chemistry also reveal that Eocene and Miocene basic vol-canic rocks were affected by more pronounced crustal contamination than the basaltic volcanic rocks of Mio-Pliocene and Quaternary.Furthermore,the integrated lithophile element and HSE compositions of these basic volcanic rocks,together with the regional asymmetric lithospheric delamination model,reveal that the compositional variation(especially due to metasomatism)was significant temporally in the heterogeneity of the SCLM sources from which parental magmas formed during the Cenozoic era.展开更多
Knowledge graph(KG)serves as a specialized semantic network that encapsulates intricate relationships among real-world entities within a structured framework.This framework facilitates a transformation in information ...Knowledge graph(KG)serves as a specialized semantic network that encapsulates intricate relationships among real-world entities within a structured framework.This framework facilitates a transformation in information retrieval,transitioning it from mere string matching to far more sophisticated entity matching.In this transformative process,the advancement of artificial intelligence and intelligent information services is invigorated.Meanwhile,the role ofmachine learningmethod in the construction of KG is important,and these techniques have already achieved initial success.This article embarks on a comprehensive journey through the last strides in the field of KG via machine learning.With a profound amalgamation of cutting-edge research in machine learning,this article undertakes a systematical exploration of KG construction methods in three distinct phases:entity learning,ontology learning,and knowledge reasoning.Especially,a meticulous dissection of machine learningdriven algorithms is conducted,spotlighting their contributions to critical facets such as entity extraction,relation extraction,entity linking,and link prediction.Moreover,this article also provides an analysis of the unresolved challenges and emerging trajectories that beckon within the expansive application of machine learning-fueled,large-scale KG construction.展开更多
The purpose of this study is to investigate the suppression effect of a nonlinear energy sink(NES)on the wind-vortex-induced pipe vibration and explore the influence of damping,stiffness,and NES installation position ...The purpose of this study is to investigate the suppression effect of a nonlinear energy sink(NES)on the wind-vortex-induced pipe vibration and explore the influence of damping,stiffness,and NES installation position on the suppression effect.In this work,the wind-vortex-induced vibration of an elastic pipe of a deepwater jacket was studied,and vibrations were suppressed by using an NES.A van der Pol wake oscillator was used to simulate vortex-induced force,and the dynamic equation of the pipe considering the NES was established.The Galerkin method was applied to discretize the motion equation,and the vortex-induced vibration(VIV)of the pipe at reduced wind speeds was numerically analyzed.The novelty of this research is that particle swarm optimization was used to optimize the parameters of the NES to improve vibration suppression.The influence of the installation position,nonlinear stiffness,and damping parameters of the NES on vibration suppression was analyzed.Results showed that the optimized parameter combinations of the NES can effectively reduce wind-vortex-induced pipe vibration.The installation position of the NES had a significant effect on vibration suppression,and the midpoint of the pipe was the optimal NES installation position.An increase in stiffness or a 10% decrease in damping may cause vibration suppression failure.The results of this study provide some guidance for VIV suppression in deepwater jacket pipes.展开更多
Inspired by the demand of improving the riding comfort and meeting the lightweight design of the vehicle, an inerter-based X-structure nonlinear energy sink(IXNES) is proposed and applied in the half-vehicle system to...Inspired by the demand of improving the riding comfort and meeting the lightweight design of the vehicle, an inerter-based X-structure nonlinear energy sink(IXNES) is proposed and applied in the half-vehicle system to enhance the dynamic performance. The X-structure is used as a mechanism to realize the nonlinear stiffness characteristic of the NES, which can realize the flexibility, adjustability, high efficiency, and easy operation of nonlinear stiffness, and is convenient to apply in the vehicle suspension, and the inerter is applied to replacing the mass of the NES based on the mass amplification characteristic. The dynamic model of the half-vehicle system coupled with the IX-NES is established with the Lagrange theory, and the harmonic balance method(HBM) and the pseudo-arc-length method(PALM) are used to obtain the dynamic response under road harmonic excitation. The corresponding dynamic performance under road harmonic and random excitation is evaluated by six performance indices, and compared with that of the original half-vehicle system to show the benefits of the IX-NES. Furthermore, the structural parameters of the IX-NES are optimized with the genetic algorithm. The results show that for road harmonic and random excitation, using the IX-NES can greatly reduce the resonance peaks and root mean square(RMS) values of the front and rear suspension deflections and the front and rear dynamic tire loads, while the resonance peaks and RMS values of the vehicle body vertical and pitching accelerations are slightly larger.When the structural parameters of the IX-NES are optimized, the vehicle body vertical and pitching accelerations of the half-vehicle system could reduce by 2.41% and 1.16%,respectively, and the other dynamic performance indices are within the reasonable ranges.Thus, the IX-NES combines the advantages of the inerter, X-structure, and NES, which improves the dynamic performance of the half-vehicle system and provides an effective option for vibration attenuation in the vehicle engineering.展开更多
Due to the structural dependencies among concurrent events in the knowledge graph and the substantial amount of sequential correlation information carried by temporally adjacent events,we propose an Independent Recurr...Due to the structural dependencies among concurrent events in the knowledge graph and the substantial amount of sequential correlation information carried by temporally adjacent events,we propose an Independent Recurrent Temporal Graph Convolution Networks(IndRT-GCNets)framework to efficiently and accurately capture event attribute information.The framework models the knowledge graph sequences to learn the evolutionary represen-tations of entities and relations within each period.Firstly,by utilizing the temporal graph convolution module in the evolutionary representation unit,the framework captures the structural dependency relationships within the knowledge graph in each period.Meanwhile,to achieve better event representation and establish effective correlations,an independent recurrent neural network is employed to implement auto-regressive modeling.Furthermore,static attributes of entities in the entity-relation events are constrained andmerged using a static graph constraint to obtain optimal entity representations.Finally,the evolution of entity and relation representations is utilized to predict events in the next subsequent step.On multiple real-world datasets such as Freebase13(FB13),Freebase 15k(FB15K),WordNet11(WN11),WordNet18(WN18),FB15K-237,WN18RR,YAGO3-10,and Nell-995,the results of multiple evaluation indicators show that our proposed IndRT-GCNets framework outperforms most existing models on knowledge reasoning tasks,which validates the effectiveness and robustness.展开更多
基金supported by the National Key R&D Program of China(2019YFB2103202).
文摘Nowadays,ensuring thequality of networkserviceshas become increasingly vital.Experts are turning toknowledge graph technology,with a significant emphasis on entity extraction in the identification of device configurations.This research paper presents a novel entity extraction method that leverages a combination of active learning and attention mechanisms.Initially,an improved active learning approach is employed to select the most valuable unlabeled samples,which are subsequently submitted for expert labeling.This approach successfully addresses the problems of isolated points and sample redundancy within the network configuration sample set.Then the labeled samples are utilized to train the model for network configuration entity extraction.Furthermore,the multi-head self-attention of the transformer model is enhanced by introducing the Adaptive Weighting method based on the Laplace mixture distribution.This enhancement enables the transformer model to dynamically adapt its focus to words in various positions,displaying exceptional adaptability to abnormal data and further elevating the accuracy of the proposed model.Through comparisons with Random Sampling(RANDOM),Maximum Normalized Log-Probability(MNLP),Least Confidence(LC),Token Entrop(TE),and Entropy Query by Bagging(EQB),the proposed method,Entropy Query by Bagging and Maximum Influence Active Learning(EQBMIAL),achieves comparable performance with only 40% of the samples on both datasets,while other algorithms require 50% of the samples.Furthermore,the entity extraction algorithm with the Adaptive Weighted Multi-head Attention mechanism(AW-MHA)is compared with BILSTM-CRF,Mutil_Attention-Bilstm-Crf,Deep_Neural_Model_NER and BERT_Transformer,achieving precision rates of 75.98% and 98.32% on the two datasets,respectively.Statistical tests demonstrate the statistical significance and effectiveness of the proposed algorithms in this paper.
基金This research was supported by the National Key Research and Development Program[2020YFB1006302].
文摘Named entity recognition(NER)is a fundamental task of information extraction(IE),and it has attracted considerable research attention in recent years.The abundant annotated English NER datasets have significantly promoted the NER research in the English field.By contrast,much fewer efforts are made to the Chinese NER research,especially in the scientific domain,due to the scarcity of Chinese NER datasets.To alleviate this problem,we present aChinese scientificNER dataset–SciCN,which contains entity annotations of titles and abstracts derived from 3,500 scientific papers.We manually annotate a total of 62,059 entities,and these entities are classified into six types.Compared to English scientific NER datasets,SciCN has a larger scale and is more diverse,for it not only contains more paper abstracts but these abstracts are derived from more research fields.To investigate the properties of SciCN and provide baselines for future research,we adapt a number of previous state-of-theart Chinese NER models to evaluate SciCN.Experimental results show that SciCN is more challenging than other Chinese NER datasets.In addition,previous studies have proven the effectiveness of using lexicons to enhance Chinese NER models.Motivated by this fact,we provide a scientific domain-specific lexicon.Validation results demonstrate that our lexicon delivers better performance gains than lexicons of other domains.We hope that the SciCN dataset and the lexicon will enable us to benchmark the NER task regarding the Chinese scientific domain and make progress for future research.The dataset and lexicon are available at:https://github.com/yangjingla/SciCN.git.
基金supported by the Outstanding Youth Team Project of Central Universities(QNTD202308)the Ant Group through CCF-Ant Research Fund(CCF-AFSG 769498 RF20220214).
文摘Named Entity Recognition(NER)stands as a fundamental task within the field of biomedical text mining,aiming to extract specific types of entities such as genes,proteins,and diseases from complex biomedical texts and categorize them into predefined entity types.This process can provide basic support for the automatic construction of knowledge bases.In contrast to general texts,biomedical texts frequently contain numerous nested entities and local dependencies among these entities,presenting significant challenges to prevailing NER models.To address these issues,we propose a novel Chinese nested biomedical NER model based on RoBERTa and Global Pointer(RoBGP).Our model initially utilizes the RoBERTa-wwm-ext-large pretrained language model to dynamically generate word-level initial vectors.It then incorporates a Bidirectional Long Short-Term Memory network for capturing bidirectional semantic information,effectively addressing the issue of long-distance dependencies.Furthermore,the Global Pointer model is employed to comprehensively recognize all nested entities in the text.We conduct extensive experiments on the Chinese medical dataset CMeEE and the results demonstrate the superior performance of RoBGP over several baseline models.This research confirms the effectiveness of RoBGP in Chinese biomedical NER,providing reliable technical support for biomedical information extraction and knowledge base construction.
基金supported by Yunnan Provincial Major Science and Technology Special Plan Projects(Grant Nos.202202AD080003,202202AE090008,202202AD080004,202302AD080003)National Natural Science Foundation of China(Grant Nos.U21B2027,62266027,62266028,62266025)Yunnan Province Young and Middle-Aged Academic and Technical Leaders Reserve Talent Program(Grant No.202305AC160063).
文摘Chinese named entity recognition(CNER)has received widespread attention as an important task of Chinese information extraction.Most previous research has focused on individually studying flat CNER,overlapped CNER,or discontinuous CNER.However,a unified CNER is often needed in real-world scenarios.Recent studies have shown that grid tagging-based methods based on character-pair relationship classification hold great potential for achieving unified NER.Nevertheless,how to enrich Chinese character-pair grid representations and capture deeper dependencies between character pairs to improve entity recognition performance remains an unresolved challenge.In this study,we enhance the character-pair grid representation by incorporating both local and global information.Significantly,we introduce a new approach by considering the character-pair grid representation matrix as a specialized image,converting the classification of character-pair relationships into a pixel-level semantic segmentation task.We devise a U-shaped network to extract multi-scale and deeper semantic information from the grid image,allowing for a more comprehensive understanding of associative features between character pairs.This approach leads to improved accuracy in predicting their relationships,ultimately enhancing entity recognition performance.We conducted experiments on two public CNER datasets in the biomedical domain,namely CMeEE-V2 and Diakg.The results demonstrate the effectiveness of our approach,which achieves F1-score improvements of 7.29 percentage points and 1.64 percentage points compared to the current state-of-the-art(SOTA)models,respectively.
基金funded by the Double Top-Class Innovation Research Project in Cyberspace Security Enforcement Technology of People’s Public Security University of China(No.2023SYL07).
文摘In recent years,cyber attacks have been intensifying and causing great harm to individuals,companies,and countries.The mining of cyber threat intelligence(CTI)can facilitate intelligence integration and serve well in combating cyber attacks.Named Entity Recognition(NER),as a crucial component of text mining,can structure complex CTI text and aid cybersecurity professionals in effectively countering threats.However,current CTI NER research has mainly focused on studying English CTI.In the limited studies conducted on Chinese text,existing models have shown poor performance.To fully utilize the power of Chinese pre-trained language models(PLMs)and conquer the problem of lengthy infrequent English words mixing in the Chinese CTIs,we propose a residual dilated convolutional neural network(RDCNN)with a conditional random field(CRF)based on a robustly optimized bidirectional encoder representation from transformers pre-training approach with whole word masking(RoBERTa-wwm),abbreviated as RoBERTa-wwm-RDCNN-CRF.We are the first to experiment on the relevant open source dataset and achieve an F1-score of 82.35%,which exceeds the common baseline model bidirectional encoder representation from transformers(BERT)-bidirectional long short-term memory(BiLSTM)-CRF in this field by about 19.52%and exceeds the current state-of-the-art model,BERT-RDCNN-CRF,by about 3.53%.In addition,we conducted an ablation study on the encoder part of the model to verify the effectiveness of the proposed model and an in-depth investigation of the PLMs and encoder part of the model to verify the effectiveness of the proposed model.The RoBERTa-wwm-RDCNN-CRF model,the shared pre-processing,and augmentation methods can serve the subsequent fundamental tasks such as cybersecurity information extraction and knowledge graph construction,contributing to important applications in downstream tasks such as intrusion detection and advanced persistent threat(APT)attack detection.
文摘由于电离层电子密度随时间变化,且空间分布不均匀,对不同频段的无线电波产生延缓和折射,因此电离层电子密度变化是影响短波通信、卫星通信、全球导航卫星系统和其他空间通信质量的一个主要因素,本文对全球电离层电子密度(Number of electron,Ne)的预测工作对短波通信设备三维射线实时追踪定位提供必要条件。本文采用国际电离层参考模型提供的2016年电离层Ne数据,根据数据的三维空间时间序列特征,搭建了自编码器和卷积长短期记忆(Convolutional Long Short-Term Memory Network,Conv LSTM)网络组成的网络结构,在不引入地球自转周期之外任何先验知识的条件下,对Ne数据进行深度学习并实现预测,首先通过实验对比了SGD、Adagrad、Adadelta、Adam、Adamax和Nadam六种优化算法的性能,又对比了三种预测策略的均方根误差(Root Mean Square Error, RMSE),1h-to-1h预测策略的全球平均RMSE为1.0 NEU(最大值的0.4%),1h-to-24h和24h-to-24h预测策略的全球平均RMSE为6.3 NEU(2.6%)。由实验结果得出以下结论,一是Nadam优化算法更适合电离层Ne的深度学习,二是1h预测策略的性能与之前类似的电离层TEC预测工作(RMSE高于1.5 TECU,最大值的1%)相比有竞争力,但预测时间太短且对数据的实时性要求较高,三是两种24h预测策略虽能实现长期预测但性能不理想,要实现三维空间时间序列的长期高精度预测需要进一步改善神经网络、模型结构和预测策略。
基金This research was supported by the National Natural Science Foundation of China under Grant(No.42050102)the Postgraduate Education Reform Project of Jiangsu Province under Grant(No.SJCX22_0343)Also,this research was supported by Dou Wanchun Expert Workstation of Yunnan Province(No.202205AF150013).
文摘With the rapid development of information technology,the electronifi-cation of medical records has gradually become a trend.In China,the population base is huge and the supporting medical institutions are numerous,so this reality drives the conversion of paper medical records to electronic medical records.Electronic medical records are the basis for establishing a smart hospital and an important guarantee for achieving medical intelligence,and the massive amount of electronic medical record data is also an important data set for conducting research in the medical field.However,electronic medical records contain a large amount of private patient information,which must be desensitized before they are used as open resources.Therefore,to solve the above problems,data masking for Chinese electronic medical records with named entity recognition is proposed in this paper.Firstly,the text is vectorized to satisfy the required format of the model input.Secondly,since the input sentences may have a long or short length and the relationship between sentences in context is not negligible.To this end,a neural network model for named entity recognition based on bidirectional long short-term memory(BiLSTM)with conditional random fields(CRF)is constructed.Finally,the data masking operation is performed based on the named entity recog-nition results,mainly using regular expression filtering encryption and principal component analysis(PCA)word vector compression and replacement.In addi-tion,comparison experiments with the hidden markov model(HMM)model,LSTM-CRF model,and BiLSTM model are conducted in this paper.The experi-mental results show that the method used in this paper achieves 92.72%Accuracy,92.30%Recall,and 92.51%F1_score,which has higher accuracy compared with other models.
基金Project supported by the National Science Fund for Distinguished Young Scholars of China(No.12025204)the National Natural Science Foundation of China(No.12202038)。
文摘With its complex nonlinear dynamic behavior,the tristable system has shown excellent performance in areas such as energy harvesting and vibration suppression,and has attracted a lot of attention.In this paper,an asymmetric tristable design is proposed to improve the vibration suppression efficiency of nonlinear energy sinks(NESs)for the first time.The proposed asymmetric tristable NES(ATNES)is composed of a pair of oblique springs and a vertical spring.Then,the three stable states,symmetric and asymmetric,can be achieved by the adjustment of the distance and stiffness asymmetry of the oblique springs.The governing equations of a linear oscillator(LO)coupled with the ATNES are derived.The approximate analytical solution to the coupled system is obtained by the harmonic balance method(HBM)and verified numerically.The vibration suppression efficiency of three types of ATNES is compared.The results show that the asymmetric design can improve the efficiency of vibration reduction through comparing the chaotic motion of the NES oscillator between asymmetric steady states.In addition,compared with the symmetrical tristable NES(TNES),the ATNES can effectively control smaller structural vibrations.In other words,the ATNES can effectively solve the threshold problem of TNES failure to weak excitation.Therefore,this paper reveals the vibration reduction mechanism of the ATNES,and provides a pathway to expand the effective excitation amplitude range of the NES.
基金supported by the Sichuan Science and Technology Program under Grant No.2021YFQ0009。
文摘The extraction and understanding of text knowledge become increasingly crucial in the age of big data.One of the current research areas in the field of natural language processing(NLP)is how to accurately understand the text and collect accurate linguistic information because Chinese vocabulary is diverse and ambiguous.This paper mainly studies the candidate entity generation module of the entity link system.The candidate entity generation module constructs an entity reference expansion algorithm to improve the recall rate of candidate entities.In order to improve the efficiency of the connection algorithm of the entire system while ensuring the recall rate of candidate entities,we design a graph model filtering algorithm that fuses shallow semantic information to filter the list of candidate entities,and verify and analyze the efficiency of the algorithm through experiments.By analyzing the related technology of the entity linking algorithm,we study the related technology of candidate entity generation and entity disambiguation,improve the traditional entity linking algorithm,and give an innovative and practical entity linking model.The recall rate exceeds 82%,and the link accuracy rate exceeds 73%.Efficient and accurate entity linking can help machines to better understand text semantics,further promoting the development of NLP and improving the users’knowledge acquisition experience on the text.
基金supported by National Natural Science Foundation of China(No.41971244)National Natural Science Foundation of China(No.41501104)Natural Science Foundation of Chongqing Municipal Science and Technology Commission(cstc2021jcyj-msxm X0696)。
文摘The natural and economic conditions of agricultural resources vary greatly in China,especially in hilly and mountainous areas.The phenomenon of land fragmentation has become increasingly prominent,so that large-scale operations cannot be fully achieved in a short period of time,and the centralized and continuous scale of operations cannot be realized in China.In addition,with increasingly strict farmland protection and agricultural land use control systems,the issues of nongrain production and nonagricultural cultivated land use have become increasingly restricted.Thus,it is quite necessary to determine the appropriate scale of operations and the means to achieve moderately scaled operations for the new business entities.On the basis of microsurvey data for 108 new citrus-planting business entities in the modern agricultural park in the Chongqing's Jiangjin District,an area with long citrus-planting history,we measured the moderate scale of new citrus-planting business entities to maximize profit using a translog growth model.According to the projection pursuit model,we evaluated the suitability conditions of citrus planting in garden land,woodland,grassland,and general cultivated land in the study area.We then explored the potential for new moderate-scale business entities within different farming radii.The results showed that large-scale planting conditions of citrus in garden land,forest land,grassland,and general cultivated land in the study area were suitable,and the proportion of high-suitable and mediumsuitable land was 73.42%.Under the existing social and economic conditions,the moderate scale of new citrus-planting business entities in the study area was 1.8–2.7 hm^(2).In particular,its rankings from large to small were agricultural enterprises(17.19–25.78 hm^(2)),farmer cooperatives(16.88–25.33 hm^(2)),big growers and breeders(6.39–9.59 hm^(2)),and family farms(5.02–7.53 hm^(2)).In the sample of 108 households,only 47.22%of the entities achieved moderate-scale operation,of which 25%of the entities achieved a scale greater than moderate operation.However,52.78%of the entities achieved a scale of less than moderate operation.These entities would have to transfer the surrounding adjacent garden land,forest land,grassland,or general cultivated land to achieve largescale land management.The land area of the adjacent gardens in a 1000 m farming radius could meet the moderate-scale operation demand of the vast majority of new business entities.For the vast majority of new business entities,the land area of the adjacent woodland,grassland,and general cultivated land could supplement the garden land to achieve moderate-scale operation in a 500 m farming radius.If the land area with moderate suitability and high suitability is prioritized,the land area in the adjacent gardens in the 1000 m farming radius could meet the moderate-scale operation demand for the vast majority of new business entities.Within the 500 m farming radius,the vast majority of new business entities have achieved moderate-scale operations if the land area adjacent to forest land,grassland,and general arable land was supplemented by gardens;however,a few entities could not achieve moderate-scale operations.From the village perspective,gardens were prioritized.Sixteen villages had planting areas and planting suitability that exceeded the average level of the study area,accounting for 23.53%.If combined with the reserve potential of the garden land,eight villages could improve their potential,accounting for 11.76%.Therefore,the conditions of large-scale citrus planting in the study area should be further improved,and the scale expansion of new citrus-planting business entities should receive additional scientific guidance.
基金General Program of Anhui University of Finance and Economics,Research on the Optimization Mechanism of Rural Governance Structure under Common Prosperity(ACKYC22041)。
文摘The positive impact of the development of new agricultural business entities and their institutional systems on the optimization of rural governance structures can be examined from various perspectives,including the state,market,rural society,urban-rural relations,and grassroots governance structure in rural communities.The development of these new agricultural business entities has not only redefined but also restructured the power distribution and governance patterns in rural developing countries,markets,and villages.The rural social order has evolved into a ternary mutual structure governance pattern,often referred to as the“state market rural”model.This transformation has prompted adjustments in the national economic and social policy structure and management systems at both macro and micro levels.It has led to the reshaping of power dynamics,benefit distribution,and governance structures in both urban and rural areas,resulting in significant changes to the economic and social fabric of rural regions.Furthermore,the grassroots governance structure in rural society,characterized by“township governance and village governance,”is undergoing continuous development and improvement.This transition is marked by a shift towards a collaborative governance structure that encourages diverse participation.Building upon the aforementioned optimizations,the rural governance structure now exhibits new characteristics.These include a more extensive and diverse range of rural governance mechanisms,increased openness in governance processes,and a heightened synergy among various governance mechanisms.This dynamic evolution reflects a richer,more diverse,and more open approach to rural governance,fostering a stronger collaborative effort in the pursuit of effective governance.
文摘The post-collisional Cenozoic basic volcanic rocks in NE Turkey show temporal variations in whole-rock lithophile element and highly siderophile element(HSE)systematics that are mainly associated with the nature of sub-continental lithospheric mantle(SCLM)sources and parental melt generation.So far,the traditional whole-rock lithophile geochemical data of these basic volcanic rocks have provided important constraints on the nature of SCLM sources.Integrated lithophile element and HSE geochemical data of these basic volcanic rocks also reveal the heterogeneity of the SCLM source,which is principally related to variable metasomatism resulting from previous subduction(s)and post-collisional mantle-crust interactions in an extensional setting.Lithophile element geochemical features suggest that the parental magmas have derived from metasomatized spinel-to garnet-bearing SCLM sources for Eocene and Miocene basic volcanic rocks with subduction signatures whereas originated from spinel-to garnet-bearing SCLM sources for Mio-Pliocene and Plio-Quaternary basaltic volcanic rocks without the subduction signature.Lithophile element and HSE geo-chemistry also reveal that Eocene and Miocene basic vol-canic rocks were affected by more pronounced crustal contamination than the basaltic volcanic rocks of Mio-Pliocene and Quaternary.Furthermore,the integrated lithophile element and HSE compositions of these basic volcanic rocks,together with the regional asymmetric lithospheric delamination model,reveal that the compositional variation(especially due to metasomatism)was significant temporally in the heterogeneity of the SCLM sources from which parental magmas formed during the Cenozoic era.
基金supported in part by the Beijing Natural Science Foundation under Grants L211020 and M21032in part by the National Natural Science Foundation of China under Grants U1836106 and 62271045in part by the Scientific and Technological Innovation Foundation of Foshan under Grants BK21BF001 and BK20BF010。
文摘Knowledge graph(KG)serves as a specialized semantic network that encapsulates intricate relationships among real-world entities within a structured framework.This framework facilitates a transformation in information retrieval,transitioning it from mere string matching to far more sophisticated entity matching.In this transformative process,the advancement of artificial intelligence and intelligent information services is invigorated.Meanwhile,the role ofmachine learningmethod in the construction of KG is important,and these techniques have already achieved initial success.This article embarks on a comprehensive journey through the last strides in the field of KG via machine learning.With a profound amalgamation of cutting-edge research in machine learning,this article undertakes a systematical exploration of KG construction methods in three distinct phases:entity learning,ontology learning,and knowledge reasoning.Especially,a meticulous dissection of machine learningdriven algorithms is conducted,spotlighting their contributions to critical facets such as entity extraction,relation extraction,entity linking,and link prediction.Moreover,this article also provides an analysis of the unresolved challenges and emerging trajectories that beckon within the expansive application of machine learning-fueled,large-scale KG construction.
基金supported by the Tianjin Municipal Transportation Commission Project(No.2018-b2).
文摘The purpose of this study is to investigate the suppression effect of a nonlinear energy sink(NES)on the wind-vortex-induced pipe vibration and explore the influence of damping,stiffness,and NES installation position on the suppression effect.In this work,the wind-vortex-induced vibration of an elastic pipe of a deepwater jacket was studied,and vibrations were suppressed by using an NES.A van der Pol wake oscillator was used to simulate vortex-induced force,and the dynamic equation of the pipe considering the NES was established.The Galerkin method was applied to discretize the motion equation,and the vortex-induced vibration(VIV)of the pipe at reduced wind speeds was numerically analyzed.The novelty of this research is that particle swarm optimization was used to optimize the parameters of the NES to improve vibration suppression.The influence of the installation position,nonlinear stiffness,and damping parameters of the NES on vibration suppression was analyzed.Results showed that the optimized parameter combinations of the NES can effectively reduce wind-vortex-induced pipe vibration.The installation position of the NES had a significant effect on vibration suppression,and the midpoint of the pipe was the optimal NES installation position.An increase in stiffness or a 10% decrease in damping may cause vibration suppression failure.The results of this study provide some guidance for VIV suppression in deepwater jacket pipes.
基金Project supported by the National Natural Science Foundation of China (Nos. 12172153 and51805216)the China Postdoctoral Science Foundation (No. 2023M731668)the Major Project of Basic Science (Natural Science) of the Jiangsu Higher Education Institutions of China(No. 22KJA410001)。
文摘Inspired by the demand of improving the riding comfort and meeting the lightweight design of the vehicle, an inerter-based X-structure nonlinear energy sink(IXNES) is proposed and applied in the half-vehicle system to enhance the dynamic performance. The X-structure is used as a mechanism to realize the nonlinear stiffness characteristic of the NES, which can realize the flexibility, adjustability, high efficiency, and easy operation of nonlinear stiffness, and is convenient to apply in the vehicle suspension, and the inerter is applied to replacing the mass of the NES based on the mass amplification characteristic. The dynamic model of the half-vehicle system coupled with the IX-NES is established with the Lagrange theory, and the harmonic balance method(HBM) and the pseudo-arc-length method(PALM) are used to obtain the dynamic response under road harmonic excitation. The corresponding dynamic performance under road harmonic and random excitation is evaluated by six performance indices, and compared with that of the original half-vehicle system to show the benefits of the IX-NES. Furthermore, the structural parameters of the IX-NES are optimized with the genetic algorithm. The results show that for road harmonic and random excitation, using the IX-NES can greatly reduce the resonance peaks and root mean square(RMS) values of the front and rear suspension deflections and the front and rear dynamic tire loads, while the resonance peaks and RMS values of the vehicle body vertical and pitching accelerations are slightly larger.When the structural parameters of the IX-NES are optimized, the vehicle body vertical and pitching accelerations of the half-vehicle system could reduce by 2.41% and 1.16%,respectively, and the other dynamic performance indices are within the reasonable ranges.Thus, the IX-NES combines the advantages of the inerter, X-structure, and NES, which improves the dynamic performance of the half-vehicle system and provides an effective option for vibration attenuation in the vehicle engineering.
基金the National Natural Science Founda-tion of China(62062062)hosted by Gulila Altenbek.
文摘Due to the structural dependencies among concurrent events in the knowledge graph and the substantial amount of sequential correlation information carried by temporally adjacent events,we propose an Independent Recurrent Temporal Graph Convolution Networks(IndRT-GCNets)framework to efficiently and accurately capture event attribute information.The framework models the knowledge graph sequences to learn the evolutionary represen-tations of entities and relations within each period.Firstly,by utilizing the temporal graph convolution module in the evolutionary representation unit,the framework captures the structural dependency relationships within the knowledge graph in each period.Meanwhile,to achieve better event representation and establish effective correlations,an independent recurrent neural network is employed to implement auto-regressive modeling.Furthermore,static attributes of entities in the entity-relation events are constrained andmerged using a static graph constraint to obtain optimal entity representations.Finally,the evolution of entity and relation representations is utilized to predict events in the next subsequent step.On multiple real-world datasets such as Freebase13(FB13),Freebase 15k(FB15K),WordNet11(WN11),WordNet18(WN18),FB15K-237,WN18RR,YAGO3-10,and Nell-995,the results of multiple evaluation indicators show that our proposed IndRT-GCNets framework outperforms most existing models on knowledge reasoning tasks,which validates the effectiveness and robustness.