A simulation model was proposed to investigate the relationship between train delays and passenger delays and to predict the dynamic passenger distribution in a large-scale rail transit network. It was assumed that th...A simulation model was proposed to investigate the relationship between train delays and passenger delays and to predict the dynamic passenger distribution in a large-scale rail transit network. It was assumed that the time varying original-destination demand and passenger path choice probability were given. Passengers were assumed not to change their destinations and travel paths after delay occurs. CapaciW constraints of train and queue rules of alighting and boarding were taken into account. By using the time-driven simulation, the states of passengers, trains and other facilities in the network were updated every time step. The proposed methodology was also tested in a real network, for demonstration. The results reveal that short train delay does not necessarily result in passenger delays, while, on the contrary, some passengers may get benefits from the short delay. However, large initial train delay may result in not only knock-on train and passenger delays along the same line, but also the passenger delays across the entire rail transit network.展开更多
Simulating large-scale and complex systems is commonly considered a difficult and time-consuming task. In this paper, we propose a partial simulation way to speed up the simulation with real time demands. It is based ...Simulating large-scale and complex systems is commonly considered a difficult and time-consuming task. In this paper, we propose a partial simulation way to speed up the simulation with real time demands. It is based on the idea that a train traffic diagram is expressed in a network, and through calculating the maximal long path in the network the simulation is done, but only within a particular partial area.Upon this, we let it become a problem oriented simulation. The simulation could be started at any time,from any trains or at any stations and stopped as the same way according to the problem to be concerned.We can use this kind of simulation to analyse or confirm the correctness of traffic schedule at a high speed to meet the real time demands.展开更多
基于“预训练+微调”范式的实体关系联合抽取方法依赖大规模标注数据,在数据标注难度大、成本高的中文古籍小样本场景下微调效率低,抽取性能不佳;中文古籍中普遍存在实体嵌套和关系重叠的问题,限制了实体关系联合抽取的效果;管道式抽取...基于“预训练+微调”范式的实体关系联合抽取方法依赖大规模标注数据,在数据标注难度大、成本高的中文古籍小样本场景下微调效率低,抽取性能不佳;中文古籍中普遍存在实体嵌套和关系重叠的问题,限制了实体关系联合抽取的效果;管道式抽取方法存在错误传播问题,影响抽取效果。针对以上问题,提出一种基于提示学习和全局指针网络的中文古籍实体关系联合抽取方法。首先,利用区间抽取式阅读理解的提示学习方法对预训练语言模型(PLM)注入领域知识以统一预训练和微调的优化目标,并对输入句子进行编码表示;其次,使用全局指针网络分别对主、客实体边界和不同关系下的主、客实体边界进行预测和联合解码,对齐成实体关系三元组,并构建了PTBG(Prompt Tuned BERT with Global pointer)模型,解决实体嵌套和关系重叠问题,同时避免了管道式解码的错误传播问题;最后,在上述工作基础上分析了不同提示模板对抽取性能的影响。在《史记》数据集上进行实验的结果表明,相较于注入领域知识前后的OneRel模型,PTBG模型所取得的F1值分别提升了1.64和1.97个百分点。可见,PTBG模型能更好地对中文古籍实体关系进行联合抽取,为低资源的小样本深度学习场景提供了新的研究思路与方法。展开更多
基金Project(51008229)supported by the National Natural Science Foundation of ChinaProject supported by Key Laboratory of Road and Traffic Engineering of Tongji University,China
文摘A simulation model was proposed to investigate the relationship between train delays and passenger delays and to predict the dynamic passenger distribution in a large-scale rail transit network. It was assumed that the time varying original-destination demand and passenger path choice probability were given. Passengers were assumed not to change their destinations and travel paths after delay occurs. CapaciW constraints of train and queue rules of alighting and boarding were taken into account. By using the time-driven simulation, the states of passengers, trains and other facilities in the network were updated every time step. The proposed methodology was also tested in a real network, for demonstration. The results reveal that short train delay does not necessarily result in passenger delays, while, on the contrary, some passengers may get benefits from the short delay. However, large initial train delay may result in not only knock-on train and passenger delays along the same line, but also the passenger delays across the entire rail transit network.
文摘Simulating large-scale and complex systems is commonly considered a difficult and time-consuming task. In this paper, we propose a partial simulation way to speed up the simulation with real time demands. It is based on the idea that a train traffic diagram is expressed in a network, and through calculating the maximal long path in the network the simulation is done, but only within a particular partial area.Upon this, we let it become a problem oriented simulation. The simulation could be started at any time,from any trains or at any stations and stopped as the same way according to the problem to be concerned.We can use this kind of simulation to analyse or confirm the correctness of traffic schedule at a high speed to meet the real time demands.
文摘基于“预训练+微调”范式的实体关系联合抽取方法依赖大规模标注数据,在数据标注难度大、成本高的中文古籍小样本场景下微调效率低,抽取性能不佳;中文古籍中普遍存在实体嵌套和关系重叠的问题,限制了实体关系联合抽取的效果;管道式抽取方法存在错误传播问题,影响抽取效果。针对以上问题,提出一种基于提示学习和全局指针网络的中文古籍实体关系联合抽取方法。首先,利用区间抽取式阅读理解的提示学习方法对预训练语言模型(PLM)注入领域知识以统一预训练和微调的优化目标,并对输入句子进行编码表示;其次,使用全局指针网络分别对主、客实体边界和不同关系下的主、客实体边界进行预测和联合解码,对齐成实体关系三元组,并构建了PTBG(Prompt Tuned BERT with Global pointer)模型,解决实体嵌套和关系重叠问题,同时避免了管道式解码的错误传播问题;最后,在上述工作基础上分析了不同提示模板对抽取性能的影响。在《史记》数据集上进行实验的结果表明,相较于注入领域知识前后的OneRel模型,PTBG模型所取得的F1值分别提升了1.64和1.97个百分点。可见,PTBG模型能更好地对中文古籍实体关系进行联合抽取,为低资源的小样本深度学习场景提供了新的研究思路与方法。
