检索结果-南通市图书馆

Hashtag Recommendation Using LSTM Networks with Self-Attention

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2019年第9期61卷 1261-1269页

作者： yatian shen Yan Li Jun Sun Wenke Ding Xianjin Shi Lei Zhang Xiajiong shen Jing He School of Computer and Information Engineering Henan UniversityKaifeng475000China The Corporate and Investment Bank TechnologyCanary WharfLondonE145JPUK

On Twitter,people often use hashtags to mark the subject of a *** have specific themes or content that are easy for people to *** the increase in the number of tweets,how to automatically recommend hashtags for tweets has received wide *** previous hashtag recommendation methods were to convert the task into a multi-class classification ***,these methods can only recommend hashtags that appeared in historical information,and cannot recommend the new *** this work,we extend the self-attention mechanism to turn the hashtag recommendation task into a sequence labeling *** train and evaluate the proposed method,we used the real tweet data which is collected from *** results show that the proposed method can be significantly better than the most advanced *** with the state-of-the-art methods,the accuracy of our method has been increased 4%.

关键词： Hashtags recommendation self-attention neural networks sequence labeling

A Deletable and Modifiable Blockchain Scheme Based on Record Verification Trees and the Multisignature Mechanism

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2021年第7期128卷 223-245页

作者： Daojun Han Jinyu Chen Lei Zhang yatian shen Yihua Gao Xueheng Wang Henan Key Laboratory of Big Data Analysis and Processing Henan UniversityKaifeng47500China Institute of Data and Knowledge Engineering Henan UniversityKaifeng47500China School of Computer and Information Engineering Henan UniversityKaifeng47500China Department of Network and Finance Agricultural Bank of China Limited Xuchang BranchXuchang461000China School of Software Pingdingshan UniversityPingdingshan467000China Department of Network and Finance Agricultural Bank of China Limited Nanyang BranchNanyang473000China

As one of the most valuable technologies,blockchains have received extensive attention from researchers and industry circles and are widely applied in various ***,data on a blockchain cannot be *** a result,it is impossible to clean invalid and sensitive data and correct erroneous ***,to a certain extent,hinders the application of blockchains in supply chains and Internet of *** address this problem,this study presents a deletable and modifiable blockchain scheme(DMBlockChain)based on record verification trees(RVTrees)and the multisignature scheme.(1)In this scheme,an RVTree structure is designed and added to the block *** RVTree can not only ensure that a record is true and valid but,owing to its unique binary structure,also verify whether modification and deletion requests are valid.(2)In DMBlockChain,the multisignature mechanism is also *** mechanism requires the stakeholders’signatures for each modification or deletion request and thus ensures that a record will not be modified arbitrarily.A user’s request is deemed valid only if it is dually verified by the RVTree and the multisignature *** analysis finds that DMBlockChain can provide a secure and valid means for modifying and deleting records in a block while ensuring the integrity of the block and that DMBlockChain can effectively save space in some scenarios that require frequent records modification.

关键词： Blockchain record verification trees multisignature DMBlockChain

使用深度长短时记忆模型对于评价词和评价对象的联合抽取

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

中文信息学报 2018年第2期32卷 110-119页

作者：沈亚田黄萱菁曹均阔复旦大学计算机科学技术学院上海201203 海南师范大学信息科学技术学院海南海口570100

评价词和评价对象抽取在意见挖掘中是一个重要的任务,我们在句子级评价词和评价对象联合抽取任务上研究了长短时记忆(long short-term memory)神经网络模型的几种变种应用。长短时记忆神经网络模型是一种循环神经网络模型,该模型使用长... 详细信息

评价词和评价对象抽取在意见挖掘中是一个重要的任务,我们在句子级评价词和评价对象联合抽取任务上研究了长短时记忆(long short-term memory)神经网络模型的几种变种应用。长短时记忆神经网络模型是一种循环神经网络模型,该模型使用长短时记忆模型单元作为循环神经网络的记忆单元,它能够获得更多的长距离上下文信息,同时避免了普通循环神经网络的梯度消失和梯度爆炸的问题。我们对比了传统的方法,实验结果证明长短时记忆神经网络模型优于以前的方法,在细粒度评价词和评价对象的联合抽取中达到更好的性能。

关键词：循环神经网络长短时记忆模型评价词与评价对象联合抽取深度学习序列标注

四川盆地PT1井区海相地层工程地质特征研究

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

钻采工艺 2024年第3期47卷 115-121页

作者：范宇郭建华沈欣宇王秋彤李亚天唐宜家李皋中国石油西南油气田公司油气藏地质及开发工程全国重点实验室西南石油大学

PT1井区超深海相地层钻井过程中气侵和井漏频发,严重制约该区块的安全高效开发。文章基于有效应力法,利用最优化方法开展了海相地层孔隙压力评价;结合断层和裂缝发育程度探讨了PT1井区气侵和井漏纵横向分布特征。结果表明,受异常高压和... 详细信息

PT1井区超深海相地层钻井过程中气侵和井漏频发,严重制约该区块的安全高效开发。文章基于有效应力法,利用最优化方法开展了海相地层孔隙压力评价;结合断层和裂缝发育程度探讨了PT1井区气侵和井漏纵横向分布特征。结果表明,受异常高压和裂缝发育程度的双重影响,气侵主要发生在龙潭组,占比达40%;井漏主要发生在筇竹寺组和灯影组,占总漏失量的80%以上。气侵与异常压力、裂缝发育程度、断裂情况等因素有关,异常高压是气侵内在的原因,而断裂带和发育的构造裂缝为气侵提供了必要条件。井漏层位和漏失量主要受一、二级大型断裂带控制,不同断裂带的影响具有显著的叠加效应,断裂带附近发育的构造裂缝为漏失提供了通道,距离F I7断裂越近越易出现漏失,且漏失量越大。研究成果为PT1井区钻井过程井下复杂预防和处理提供了科学支撑。

关键词： PT1井区气侵井漏孔隙压力裂缝发育程度断裂带

Joint Self-Attention Based Neural Networks for Semantic Relation Extraction

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Journal of Information Hiding and Privacy Protection 2019年第2期1卷 69-75页

作者： Jun Sun Yan Li yatian shen Wenke Ding Xianjin Shi Lei Zhang Xiajiong shen Jing He School of Computer and Information Engineering Henan UniversityKaifeng475000China The Corporate and Investment Bank TechnologyJ.P.Morgan Chase N.A.25 Bank StCanary WharfLondonE145JPUnited Kingdom

Relation extraction is an important task in NLP ***,some models often fail in capturing Long-distance dependence on semantics,and the interaction between semantics of two entities is *** this paper,we propose a novel neural network model for semantic relation classification called joint self-attention bi-LSTM(SA-Bi-LSTM)to model the internal structure of the sentence to obtain the importance of each word of the sentence without relying on additional information,and capture Long-distance dependence on *** conduct experiments using the SemEval-2010 Task 8 *** experiments and the results demonstrated that the proposed method is effective against relation classification,which can obtain state-ofthe-art classification accuracy just with minimal feature engineering.

关键词： Self-attention relation extraction neural networks