Big data have the characteristics of enormous volume,high velocity,diversity,value-sparsity,and uncertainty,which lead the knowledge learning from them full of challenges.With the emergence of crowdsourcing,versatile ...Big data have the characteristics of enormous volume,high velocity,diversity,value-sparsity,and uncertainty,which lead the knowledge learning from them full of challenges.With the emergence of crowdsourcing,versatile information can be obtained on-demand so that the wisdom of crowds is easily involved to facilitate the knowledge learning process.During the past thirteen years,researchers in the AI community made great efforts to remove the obstacles in the field of learning from crowds.This concentrated survey paper comprehensively reviews the technical progress in crowdsourcing learning from a systematic perspective that includes three dimensions of data,models,and learning processes.In addition to reviewing existing important work,the paper places a particular emphasis on providing some promising blueprints on each dimension as well as discussing the lessons learned from our past research work,which will light up the way for new researchers and encourage them to pursue new contributions.展开更多
A composite random variable is a product (or sum of products) of statistically distributed quantities. Such a variable can represent the solution to a multi-factor quantitative problem submitted to a large, diverse, i...A composite random variable is a product (or sum of products) of statistically distributed quantities. Such a variable can represent the solution to a multi-factor quantitative problem submitted to a large, diverse, independent, anonymous group of non-expert respondents (the “crowd”). The objective of this research is to examine the statistical distribution of solutions from a large crowd to a quantitative problem involving image analysis and object counting. Theoretical analysis by the author, covering a range of conditions and types of factor variables, predicts that composite random variables are distributed log-normally to an excellent approximation. If the factors in a problem are themselves distributed log-normally, then their product is rigorously log-normal. A crowdsourcing experiment devised by the author and implemented with the assistance of a BBC (British Broadcasting Corporation) television show, yielded a sample of approximately 2000 responses consistent with a log-normal distribution. The sample mean was within ~12% of the true count. However, a Monte Carlo simulation (MCS) of the experiment, employing either normal or log-normal random variables as factors to model the processes by which a crowd of 1 million might arrive at their estimates, resulted in a visually perfect log-normal distribution with a mean response within ~5% of the true count. The results of this research suggest that a well-modeled MCS, by simulating a sample of responses from a large, rational, and incentivized crowd, can provide a more accurate solution to a quantitative problem than might be attainable by direct sampling of a smaller crowd or an uninformed crowd, irrespective of size, that guesses randomly.展开更多
Background:Crowdfunding has risen rapidly as a way of raising funds to support projects such as art projects,charity projects,and new ventures.It is very important to understand how crowds in the crowdfunding market a...Background:Crowdfunding has risen rapidly as a way of raising funds to support projects such as art projects,charity projects,and new ventures.It is very important to understand how crowds in the crowdfunding market are organized to carry out various activities.This study documents and compares two crowd designs for crowdfunding,namely pure crowds,where all crowd members participate as equals,and hybrid crowds,where crowd members are led by an expert investor.The hybrid design is rarely studied in the crowdfunding literature despite its large presence in equity crowdfunding.Methods:We examine industry practices from various countries in terms of crowd designs,review relevant literature on this topic,and develop a conceptual framework for choosing between pure and hybrid crowds.Results:We identify several inefficiencies of pure crowds in crowdfunding platforms and discuss the advantages of hybrid crowds.We then develop a conceptual framework that illustrates the factors for choosing between pure and hybrid crowds.Finally,we discuss the issue of how to manage and regulate lead investors in hybrid crowds.Conclusions:Pure crowds have several shortcomings that could be mitigated by a hybrid crowd design,especially when the proposed project suffers from greater risks,a high degree of information asymmetry,concerns about information leakage,and a high cost of managing the crowds.But for the hybrid crowd to work well,one must carefully design mechanisms for lead investor selection,compensation,and discipline.Our study contributes to the crowdfunding literature and to crowdfunding practice in multiple ways.展开更多
Zhou Dan, an articulate lawyer, led a semi-secret life until recently when he was invited to give a talk to the Homosexual Studies class at Fudan University in Shanghai.
In high-density gatherings,crowd disasters frequently occur despite all the safety measures.Timely detection of congestion in human crowds using automated analysis of video footage can prevent crowd disasters.Recent w...In high-density gatherings,crowd disasters frequently occur despite all the safety measures.Timely detection of congestion in human crowds using automated analysis of video footage can prevent crowd disasters.Recent work on the prevention of crowd disasters has been based on manual analysis of video footage.Some methods also measure crowd congestion by estimating crowd density.However,crowd density alone cannot provide reliable information about congestion.This paper proposes a deep learning framework for automated crowd congestion detection that leverages pedestrian trajectories.The proposed framework divided the input video into several temporal segments.We then extracted dense trajectories from each temporal segment and converted these into a spatio-temporal image without losing information.A classification model based on convolutional neural networks was then trained using spatio-temporal images.Next,we generated a score map by encoding each point trajectory with its respective class score.After this,we obtained the congested regions by employing the non-maximum suppression method on the score map.Finally,we demonstrated the proposed framework’s effectiveness by performing a series of experiments on challenging video sequences.展开更多
Crowding is a reality in which we find ourselves involved daily. The crowd produced in a traffic jam is a dynamic entity inwhich the application of physics, mathematics and biology can provide practical help to unders...Crowding is a reality in which we find ourselves involved daily. The crowd produced in a traffic jam is a dynamic entity inwhich the application of physics, mathematics and biology can provide practical help to understand how and why this problematicsituation occurs and what solutions can be found to resolve it. In the mobility education project ＂SicuraMENTE＂, we carried out anexperiment on the conduct of a crowd by simulating a situation of intense city traffic. Taking a cue from an experimental situationproposed in traffic physics, we have verified that a route with limited access can generate a traffic jam and the crowd of pedestrianscan be mitigated by forcing the crowd to use dedicated streets. Taking the outgoing time of the student crowd from the outlet road,with and without the presence of an obstacle in the middle of the roadway, it was found that the traffic jam is resolved more quicklyin the situation with an obstacle because the flow of people is divided into two separate channels, reducing the probability that twoindividuals are close and that this creates an obstruction （example of counter-intuitive physical principle）. We also verified that thespeed of the elements of the crowd influences the formation of traffic jams, which are on average more likely in the case of higherspeed. These important aspects in the design of road networks and transport infrastructure have made students reason on physics＇topics, but also on the correct conduct in traffic. The multidisciplinary approach in education for safe and sustainable mobility, soinnovative in Italy, turned out to be effective in terms of teaching in the frame of a mix of academic disciplines, in which road safetyeducation has become the context and the goal.展开更多
The huge number of pilgrims to the holy Mecca in the Hajj needs high awareness of crowd safety management. The stoning of the Jamarat, which is one of the rituals of the Hajj, undergoes the most dangerous crowd moveme...The huge number of pilgrims to the holy Mecca in the Hajj needs high awareness of crowd safety management. The stoning of the Jamarat, which is one of the rituals of the Hajj, undergoes the most dangerous crowd movements where fatal accidents occurred. This work investigates some problems related with the crowd dynamics when stoning the Jamarat pillars and gives some solutions. The main idea of this research is to suppose that the crowd dynamics is assimilated to fluid movement under certain conditions. Numerical simulation using a computational fluid dynamics program is used to solve Navier-Stokes equations governing the mechanics of homogeneous and incompressible fluid in a domain similar to the Jamarat Bridge from the entrance to the middle Jamarah. Some solutions are proposed inspired by the flow solutions to better manage crowd movements in the Jamarat Bridge and eventually in other similar dynamics events like sporting events.展开更多
Emergency departments(EDs)are among the busiest hospital units,where visibility is crucial for surveillance,collaboration,and wayfinding.In China,high ED patient volumes lead to crowded corridors,significantly impacti...Emergency departments(EDs)are among the busiest hospital units,where visibility is crucial for surveillance,collaboration,and wayfinding.In China,high ED patient volumes lead to crowded corridors,significantly impacting visibility.However,current visibility assessment methods focus on static obstructions such as walls and columns,neglecting crowd obstruction,and underestimating visual impediments.To address this research gap,this study proposes a novel visibility assessment method combining agent-based simulation and space syntax analysis.Based on peak-hour behavioral data from Hospital R’s ED,crowd movement was simulated across six ED plans with different corridor layouts.The simulated crowd positions were then treated as visual obstacles,and space syntax theory was applied to evaluate visibility under crowd obstruction at various timesteps.The findings reveal that:1)ring corridor layouts reduce patient backtracking and corridor congestion;2)ring corridor layouts facilitate better natural surveillance of the overall space by medical staff with crowd obstruction considered;3)ring corridor layouts enable medical staff to see each other more easily,supporting team collaboration,regardless of crowd obstruction;4)simple layouts perform best in terms of wayfinding,irrespective of crowd obstruction.This study presents the first comprehensive quantitative assessment of ED visibility under crowd obstruction,with potential applications in other crowded public spaces.展开更多
When a crowdsourcing approach is used to assist the classification of a set of items,the main objective is to classify this set of items by aggregating the worker-provided labels.A secondary objective is to assess the...When a crowdsourcing approach is used to assist the classification of a set of items,the main objective is to classify this set of items by aggregating the worker-provided labels.A secondary objective is to assess the workers’skill levels in this process.A classical model that achieves both objectives is the famous Dawid-Skene model.In this paper,we consider a third objective in this context,namely,to learn a classifier that is capable of labelling future items without further assistance of crowd workers.By extending the DawidSkene model to include the item features into consideration,we develop a Classification-Oriented Dawid Skene(CODS)model,which achieves the three objectives simultaneously.The effectiveness of CODS on this three dimensions of the problem space is demonstrated experimentally.展开更多
I had a three-day holiday for this May Day.So I went to the coastal city Qingdao.It was sunny and hot.So I had great fun playing in the water.In the afternoon,I went shopping.But there were too many people in the mark...I had a three-day holiday for this May Day.So I went to the coastal city Qingdao.It was sunny and hot.So I had great fun playing in the water.In the afternoon,I went shopping.But there were too many people in the market.I didn’t really enjoy it.展开更多
This summer holiday,my older brother and I paid a visit to my grandparents by bus.The bus was crowded,but everyone seemed to be friendly and helpful.We enjoyed beautiful views all the way.At first,we saw one hill afte...This summer holiday,my older brother and I paid a visit to my grandparents by bus.The bus was crowded,but everyone seemed to be friendly and helpful.We enjoyed beautiful views all the way.At first,we saw one hill after another.展开更多
研究探讨了对话式阅读中的CROWD提示策略(完成性、回忆性、开放性、问题性、间距性)如何促进4~5岁幼儿心理理论(Theory of Mind, ToM)的发展。首先基于改进的卡西迪分析框架,确立了“心理状态的语言表征”和“错误信念的表征”两大标准...研究探讨了对话式阅读中的CROWD提示策略(完成性、回忆性、开放性、问题性、间距性)如何促进4~5岁幼儿心理理论(Theory of Mind, ToM)的发展。首先基于改进的卡西迪分析框架,确立了“心理状态的语言表征”和“错误信念的表征”两大标准,并以此为依据选取了7本富含心理理论内容的图画书作为分析样本。随后系统地剖析了五种CROWD提示策略在这些图画书阅读中的具体应用路径,通过列举实例说明了每种策略如何引导幼儿识别和推断角色的信念、意图、情绪等心理状态。最后得出结论:CROWD策略通过构建“行为–心理结果”的因果链条、训练双重表征能力、促进反事实推理等方式,能够有效提升幼儿的心理理论能力,并通过认知与情感的双向建构,实现对话式阅读教育效能的最优化。展开更多
Soft rock is one of the common geological conditions in coal mine underground water reservoir engineering.The cross-scale correlation analysis of water erosion soft lithology deterioration is very important for the sa...Soft rock is one of the common geological conditions in coal mine underground water reservoir engineering.The cross-scale correlation analysis of water erosion soft lithology deterioration is very important for the safety and stability of coal mine underground reservoir(CMUR)engineering.To address the issues of grain crowding and segmentation difficulties in cross-scale corelation analysis,as well as the limitations of traditional etching methods,this study proposes an image grain segmentation method based on deep learning algorithms,utilizing scanning electron microscopy and image process-ing techniques.The method successfully segments crowded grains and eliminates the interference from misplaced particles.In addition,indoor uniaxial compression tests were conducted to obtain the mechanical properties of sandstone samples with different water content.By quantitatively characterizing the macroscopic and microscopic deterioration degree of red sandstone samples with different water contents,the relationship between the strength changes of rock samples and the pet-rographic parameters such as grain size and grain shape is analyzed,and the influence law of soft lithology deterioration in CMUR engineering is revealed.The results indicate:(1)Water significantly weakens the mechanical properties and stability of soft rock.With increasing water content,the strength of sandstone samples continuously decreases,and the failure mode transitions from brittle to ductile failure.(2)The deterioration of micro-micro structures is the main cause of the decrease in mechanical properties of water-eroded soft rock.Grain size,grain area,and aspect ratio are negatively correlated with water content,indicating that hydrophilic minerals in soft rock dissolve under the action of water,leading to rock damage.(3)Grain size,area,and aspect ratio can serve as significant indicators for quantifying the strength changes of water-eroded soft rock.The research findings can be applied to stability assessment and disaster prevention in CMUR engineering.展开更多
In a crowd density estimation dataset,the annotation of crowd locations is an extremely laborious task,and they are not taken into the evaluation metrics.In this paper,we aim to reduce the annotation cost of crowd dat...In a crowd density estimation dataset,the annotation of crowd locations is an extremely laborious task,and they are not taken into the evaluation metrics.In this paper,we aim to reduce the annotation cost of crowd datasets,and propose a crowd density estimation method based on weakly-supervised learning,in the absence of crowd position supervision information,which directly reduces the number of crowds by using the number of pedestrians in the image as the supervised information.For this purpose,we design a new training method,which exploits the correlation between global and local image features by incremental learning to train the network.Specifically,we design a parent-child network(PC-Net)focusing on the global and local image respectively,and propose a linear feature calibration structure to train the PC-Net simultaneously,and the child network learns feature transfer factors and feature bias weights,and uses the transfer factors and bias weights to linearly feature calibrate the features extracted from the Parent network,to improve the convergence of the network by using local features hidden in the crowd images.In addition,we use the pyramid vision transformer as the backbone of the PC-Net to extract crowd features at different levels,and design a global-local feature loss function(L2).We combine it with a crowd counting loss(LC)to enhance the sensitivity of the network to crowd features during the training process,which effectively improves the accuracy of crowd density estimation.The experimental results show that the PC-Net significantly reduces the gap between fullysupervised and weakly-supervised crowd density estimation,and outperforms the comparison methods on five datasets of Shanghai Tech Part A,ShanghaiTech Part B,UCF_CC_50,UCF_QNRF and JHU-CROWD++.展开更多
AIM:To compare superficial and deep vascular properties of optic discs between crowded discs and controls using optical coherence tomography angiography(OCT-A).METHODS:Thirty patients with crowded discs,and 47 control...AIM:To compare superficial and deep vascular properties of optic discs between crowded discs and controls using optical coherence tomography angiography(OCT-A).METHODS:Thirty patients with crowded discs,and 47 control subjects were enrolled in the study.One eye of each individual was included and OCT-A scans of optic discs were obtained in a 4.5×4.5 mm^(2) rectangular area.Radial peripapillary capillary(RPC)density,peripapillary retinal nerve fiber layer(pRNFL)thickness,cup volume,rim area,disc area,cup-to-disc(c/d)area ratio,and vertical c/d ratio were obtained automatically using device software.Automated parapapillary choroidal microvasculature(PPCMv)density was calculated using MATLAB software.When the vertical c/d ratio of the optic disc was absent or small cup,it was considered as a crowded disc.RESULTS:The mean signal strength index of OCT-A images was similar between the crowded discs and control eyes(P=0.740).There was no difference in pRNFL between the two groups(P=0.102).There were no differences in RPC density in whole image(P=0.826)and peripapillary region(P=0.923),but inside disc RPC density was higher in crowded optic discs(P=0.003).The PPCMv density in the inner-hemisuperior region was also lower in crowded discs(P=0.026).The pRNFL thickness was positively correlated with peripapillary RPC density(r=0.498,P<0.001).The inside disc RPC density was negatively correlated with c/d area ratio(r=-0.341,P=0.002).CONCLUSION:The higher inside disc RPC density and lower inner-hemisuperior PPCMv density are found in eyes with crowded optic discs.展开更多
基金supported by the National Key Research and Development Program of China(2018AAA0102002)the National Natural Science Foundation of China(62076130,91846104).
文摘Big data have the characteristics of enormous volume,high velocity,diversity,value-sparsity,and uncertainty,which lead the knowledge learning from them full of challenges.With the emergence of crowdsourcing,versatile information can be obtained on-demand so that the wisdom of crowds is easily involved to facilitate the knowledge learning process.During the past thirteen years,researchers in the AI community made great efforts to remove the obstacles in the field of learning from crowds.This concentrated survey paper comprehensively reviews the technical progress in crowdsourcing learning from a systematic perspective that includes three dimensions of data,models,and learning processes.In addition to reviewing existing important work,the paper places a particular emphasis on providing some promising blueprints on each dimension as well as discussing the lessons learned from our past research work,which will light up the way for new researchers and encourage them to pursue new contributions.
文摘A composite random variable is a product (or sum of products) of statistically distributed quantities. Such a variable can represent the solution to a multi-factor quantitative problem submitted to a large, diverse, independent, anonymous group of non-expert respondents (the “crowd”). The objective of this research is to examine the statistical distribution of solutions from a large crowd to a quantitative problem involving image analysis and object counting. Theoretical analysis by the author, covering a range of conditions and types of factor variables, predicts that composite random variables are distributed log-normally to an excellent approximation. If the factors in a problem are themselves distributed log-normally, then their product is rigorously log-normal. A crowdsourcing experiment devised by the author and implemented with the assistance of a BBC (British Broadcasting Corporation) television show, yielded a sample of approximately 2000 responses consistent with a log-normal distribution. The sample mean was within ~12% of the true count. However, a Monte Carlo simulation (MCS) of the experiment, employing either normal or log-normal random variables as factors to model the processes by which a crowd of 1 million might arrive at their estimates, resulted in a visually perfect log-normal distribution with a mean response within ~5% of the true count. The results of this research suggest that a well-modeled MCS, by simulating a sample of responses from a large, rational, and incentivized crowd, can provide a more accurate solution to a quantitative problem than might be attainable by direct sampling of a smaller crowd or an uninformed crowd, irrespective of size, that guesses randomly.
文摘Background:Crowdfunding has risen rapidly as a way of raising funds to support projects such as art projects,charity projects,and new ventures.It is very important to understand how crowds in the crowdfunding market are organized to carry out various activities.This study documents and compares two crowd designs for crowdfunding,namely pure crowds,where all crowd members participate as equals,and hybrid crowds,where crowd members are led by an expert investor.The hybrid design is rarely studied in the crowdfunding literature despite its large presence in equity crowdfunding.Methods:We examine industry practices from various countries in terms of crowd designs,review relevant literature on this topic,and develop a conceptual framework for choosing between pure and hybrid crowds.Results:We identify several inefficiencies of pure crowds in crowdfunding platforms and discuss the advantages of hybrid crowds.We then develop a conceptual framework that illustrates the factors for choosing between pure and hybrid crowds.Finally,we discuss the issue of how to manage and regulate lead investors in hybrid crowds.Conclusions:Pure crowds have several shortcomings that could be mitigated by a hybrid crowd design,especially when the proposed project suffers from greater risks,a high degree of information asymmetry,concerns about information leakage,and a high cost of managing the crowds.But for the hybrid crowd to work well,one must carefully design mechanisms for lead investor selection,compensation,and discipline.Our study contributes to the crowdfunding literature and to crowdfunding practice in multiple ways.
文摘Zhou Dan, an articulate lawyer, led a semi-secret life until recently when he was invited to give a talk to the Homosexual Studies class at Fudan University in Shanghai.
基金supported by the Ministry of Education in Saudi Arabia(Grant Number 0909).
文摘In high-density gatherings,crowd disasters frequently occur despite all the safety measures.Timely detection of congestion in human crowds using automated analysis of video footage can prevent crowd disasters.Recent work on the prevention of crowd disasters has been based on manual analysis of video footage.Some methods also measure crowd congestion by estimating crowd density.However,crowd density alone cannot provide reliable information about congestion.This paper proposes a deep learning framework for automated crowd congestion detection that leverages pedestrian trajectories.The proposed framework divided the input video into several temporal segments.We then extracted dense trajectories from each temporal segment and converted these into a spatio-temporal image without losing information.A classification model based on convolutional neural networks was then trained using spatio-temporal images.Next,we generated a score map by encoding each point trajectory with its respective class score.After this,we obtained the congested regions by employing the non-maximum suppression method on the score map.Finally,we demonstrated the proposed framework’s effectiveness by performing a series of experiments on challenging video sequences.
文摘Crowding is a reality in which we find ourselves involved daily. The crowd produced in a traffic jam is a dynamic entity inwhich the application of physics, mathematics and biology can provide practical help to understand how and why this problematicsituation occurs and what solutions can be found to resolve it. In the mobility education project ＂SicuraMENTE＂, we carried out anexperiment on the conduct of a crowd by simulating a situation of intense city traffic. Taking a cue from an experimental situationproposed in traffic physics, we have verified that a route with limited access can generate a traffic jam and the crowd of pedestrianscan be mitigated by forcing the crowd to use dedicated streets. Taking the outgoing time of the student crowd from the outlet road,with and without the presence of an obstacle in the middle of the roadway, it was found that the traffic jam is resolved more quicklyin the situation with an obstacle because the flow of people is divided into two separate channels, reducing the probability that twoindividuals are close and that this creates an obstruction （example of counter-intuitive physical principle）. We also verified that thespeed of the elements of the crowd influences the formation of traffic jams, which are on average more likely in the case of higherspeed. These important aspects in the design of road networks and transport infrastructure have made students reason on physics＇topics, but also on the correct conduct in traffic. The multidisciplinary approach in education for safe and sustainable mobility, soinnovative in Italy, turned out to be effective in terms of teaching in the frame of a mix of academic disciplines, in which road safetyeducation has become the context and the goal.
文摘The huge number of pilgrims to the holy Mecca in the Hajj needs high awareness of crowd safety management. The stoning of the Jamarat, which is one of the rituals of the Hajj, undergoes the most dangerous crowd movements where fatal accidents occurred. This work investigates some problems related with the crowd dynamics when stoning the Jamarat pillars and gives some solutions. The main idea of this research is to suppose that the crowd dynamics is assimilated to fluid movement under certain conditions. Numerical simulation using a computational fluid dynamics program is used to solve Navier-Stokes equations governing the mechanics of homogeneous and incompressible fluid in a domain similar to the Jamarat Bridge from the entrance to the middle Jamarah. Some solutions are proposed inspired by the flow solutions to better manage crowd movements in the Jamarat Bridge and eventually in other similar dynamics events like sporting events.
基金funded by the National Natural Science Foundation of China under the project “Research on the Design Pattern of University-Affiliated Hospitals in the Context of Advanced Medical Care” (Grant No. 51978143) and “Research on Fire Safety Evacuation Design of Large General Hospitals based on Multi-agent Simulation” (Grant No. 52478010)
文摘Emergency departments(EDs)are among the busiest hospital units,where visibility is crucial for surveillance,collaboration,and wayfinding.In China,high ED patient volumes lead to crowded corridors,significantly impacting visibility.However,current visibility assessment methods focus on static obstructions such as walls and columns,neglecting crowd obstruction,and underestimating visual impediments.To address this research gap,this study proposes a novel visibility assessment method combining agent-based simulation and space syntax analysis.Based on peak-hour behavioral data from Hospital R’s ED,crowd movement was simulated across six ED plans with different corridor layouts.The simulated crowd positions were then treated as visual obstacles,and space syntax theory was applied to evaluate visibility under crowd obstruction at various timesteps.The findings reveal that:1)ring corridor layouts reduce patient backtracking and corridor congestion;2)ring corridor layouts facilitate better natural surveillance of the overall space by medical staff with crowd obstruction considered;3)ring corridor layouts enable medical staff to see each other more easily,supporting team collaboration,regardless of crowd obstruction;4)simple layouts perform best in terms of wayfinding,irrespective of crowd obstruction.This study presents the first comprehensive quantitative assessment of ED visibility under crowd obstruction,with potential applications in other crowded public spaces.
基金supported in part by the National Key R&D Program of China(2021ZD0110700)in part by the Fundamental Research Funds for the Central Universities,in part by the State Key Laboratory of Software Development Environmentin part by a Leverhulme Trust Research Project Grant.
文摘When a crowdsourcing approach is used to assist the classification of a set of items,the main objective is to classify this set of items by aggregating the worker-provided labels.A secondary objective is to assess the workers’skill levels in this process.A classical model that achieves both objectives is the famous Dawid-Skene model.In this paper,we consider a third objective in this context,namely,to learn a classifier that is capable of labelling future items without further assistance of crowd workers.By extending the DawidSkene model to include the item features into consideration,we develop a Classification-Oriented Dawid Skene(CODS)model,which achieves the three objectives simultaneously.The effectiveness of CODS on this three dimensions of the problem space is demonstrated experimentally.
文摘I had a three-day holiday for this May Day.So I went to the coastal city Qingdao.It was sunny and hot.So I had great fun playing in the water.In the afternoon,I went shopping.But there were too many people in the market.I didn’t really enjoy it.
文摘This summer holiday,my older brother and I paid a visit to my grandparents by bus.The bus was crowded,but everyone seemed to be friendly and helpful.We enjoyed beautiful views all the way.At first,we saw one hill after another.
文摘研究探讨了对话式阅读中的CROWD提示策略(完成性、回忆性、开放性、问题性、间距性)如何促进4~5岁幼儿心理理论(Theory of Mind, ToM)的发展。首先基于改进的卡西迪分析框架,确立了“心理状态的语言表征”和“错误信念的表征”两大标准,并以此为依据选取了7本富含心理理论内容的图画书作为分析样本。随后系统地剖析了五种CROWD提示策略在这些图画书阅读中的具体应用路径,通过列举实例说明了每种策略如何引导幼儿识别和推断角色的信念、意图、情绪等心理状态。最后得出结论:CROWD策略通过构建“行为–心理结果”的因果链条、训练双重表征能力、促进反事实推理等方式,能够有效提升幼儿的心理理论能力,并通过认知与情感的双向建构,实现对话式阅读教育效能的最优化。
基金supported by the National Natural Science Foundation of China(51774196,52304093)China Postdoctoral Science Foundation(2023M741968)Shandong Provincial Natural Science Foundation(ZR2023ME086).
文摘Soft rock is one of the common geological conditions in coal mine underground water reservoir engineering.The cross-scale correlation analysis of water erosion soft lithology deterioration is very important for the safety and stability of coal mine underground reservoir(CMUR)engineering.To address the issues of grain crowding and segmentation difficulties in cross-scale corelation analysis,as well as the limitations of traditional etching methods,this study proposes an image grain segmentation method based on deep learning algorithms,utilizing scanning electron microscopy and image process-ing techniques.The method successfully segments crowded grains and eliminates the interference from misplaced particles.In addition,indoor uniaxial compression tests were conducted to obtain the mechanical properties of sandstone samples with different water content.By quantitatively characterizing the macroscopic and microscopic deterioration degree of red sandstone samples with different water contents,the relationship between the strength changes of rock samples and the pet-rographic parameters such as grain size and grain shape is analyzed,and the influence law of soft lithology deterioration in CMUR engineering is revealed.The results indicate:(1)Water significantly weakens the mechanical properties and stability of soft rock.With increasing water content,the strength of sandstone samples continuously decreases,and the failure mode transitions from brittle to ductile failure.(2)The deterioration of micro-micro structures is the main cause of the decrease in mechanical properties of water-eroded soft rock.Grain size,grain area,and aspect ratio are negatively correlated with water content,indicating that hydrophilic minerals in soft rock dissolve under the action of water,leading to rock damage.(3)Grain size,area,and aspect ratio can serve as significant indicators for quantifying the strength changes of water-eroded soft rock.The research findings can be applied to stability assessment and disaster prevention in CMUR engineering.
基金the Humanities and Social Science Fund of the Ministry of Education of China(21YJAZH077)。
文摘In a crowd density estimation dataset,the annotation of crowd locations is an extremely laborious task,and they are not taken into the evaluation metrics.In this paper,we aim to reduce the annotation cost of crowd datasets,and propose a crowd density estimation method based on weakly-supervised learning,in the absence of crowd position supervision information,which directly reduces the number of crowds by using the number of pedestrians in the image as the supervised information.For this purpose,we design a new training method,which exploits the correlation between global and local image features by incremental learning to train the network.Specifically,we design a parent-child network(PC-Net)focusing on the global and local image respectively,and propose a linear feature calibration structure to train the PC-Net simultaneously,and the child network learns feature transfer factors and feature bias weights,and uses the transfer factors and bias weights to linearly feature calibrate the features extracted from the Parent network,to improve the convergence of the network by using local features hidden in the crowd images.In addition,we use the pyramid vision transformer as the backbone of the PC-Net to extract crowd features at different levels,and design a global-local feature loss function(L2).We combine it with a crowd counting loss(LC)to enhance the sensitivity of the network to crowd features during the training process,which effectively improves the accuracy of crowd density estimation.The experimental results show that the PC-Net significantly reduces the gap between fullysupervised and weakly-supervised crowd density estimation,and outperforms the comparison methods on five datasets of Shanghai Tech Part A,ShanghaiTech Part B,UCF_CC_50,UCF_QNRF and JHU-CROWD++.
文摘AIM:To compare superficial and deep vascular properties of optic discs between crowded discs and controls using optical coherence tomography angiography(OCT-A).METHODS:Thirty patients with crowded discs,and 47 control subjects were enrolled in the study.One eye of each individual was included and OCT-A scans of optic discs were obtained in a 4.5×4.5 mm^(2) rectangular area.Radial peripapillary capillary(RPC)density,peripapillary retinal nerve fiber layer(pRNFL)thickness,cup volume,rim area,disc area,cup-to-disc(c/d)area ratio,and vertical c/d ratio were obtained automatically using device software.Automated parapapillary choroidal microvasculature(PPCMv)density was calculated using MATLAB software.When the vertical c/d ratio of the optic disc was absent or small cup,it was considered as a crowded disc.RESULTS:The mean signal strength index of OCT-A images was similar between the crowded discs and control eyes(P=0.740).There was no difference in pRNFL between the two groups(P=0.102).There were no differences in RPC density in whole image(P=0.826)and peripapillary region(P=0.923),but inside disc RPC density was higher in crowded optic discs(P=0.003).The PPCMv density in the inner-hemisuperior region was also lower in crowded discs(P=0.026).The pRNFL thickness was positively correlated with peripapillary RPC density(r=0.498,P<0.001).The inside disc RPC density was negatively correlated with c/d area ratio(r=-0.341,P=0.002).CONCLUSION:The higher inside disc RPC density and lower inner-hemisuperior PPCMv density are found in eyes with crowded optic discs.
