|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | MSRD:多模态网络谣言检测方法显示文摘图像和文本相结合的多模态网络谣言由于更具迷惑性和煽动性,对国家安全和社会稳定的危害性更严重.目前网络谣言检测工作充分考虑了谣言中配文的文本内容而忽略了图像内容以及图像中的内嵌文本内容,因此,提出了一种基于深度神经网络针对图像、图像内嵌文本以及配文文本内容的多模态网络谣言检测方法MSRD.该方法使用VGG-19网络提取图像内容特征,使用DenseNet提取图像内嵌文本内容,使用LSTM网络提取文本内容特征,与图像特征串接后,通过完全连接层获取图像与文本共享表示的均值与方差向量,借助从高斯分布中采样的随机变量以形成重新参数化的多模态特征并作为谣言检测器的输入进行谣言检测.实验表明:该方法在Twitter和微博两大数据集上达到了68.5%和79.4%的准确率. | 刘金硕 冯阔 Jeff Z.Pan 邓娟 王丽娜 | 2020 | 计算机研究与发展2020，57，11： | 14 |
| 2 | Knowledge extraction from Chinese wiki encyclopedias显示文摘The vision of the Semantic Web is to build a 'Web of data' that enables machines to understand the semantics of information on the Web.The Linked Open Data(LOD) project encourages people and organizations to publish various open data sets as Resource Description Framework(RDF) on the Web,which promotes the development of the Semantic Web.Among various LOD datasets,DBpedia has proved a successful structured knowledge base,and has become the central interlinking-hub of the Web of data in English.However,in the Chinese language,there is little linked data published and linked to DBpedia.This hinders the structured knowledge sharing of both Chinese and cross-lingual resources.This paper deals with an approach for building a large-scale Chinese structured knowledge base from Chinese wiki resources,including Hudong and Baidu Baike.The proposed approach first builds an ontology based on the wiki category system and infoboxes,and then extracts instances from wiki articles.Using Hudong as our source,our approach builds an ontology containing 19 542 concepts and 2381 properties.802 593 instances are extracted and described using the concepts and properties in the extracted ontology and 62 679 of them are linked to equivalent instances in DBpedia.As from Baidu Baike,our approach builds an ontology containing 299 concepts,37 object properties,and 5590 data type properties.1 319 703 instances are extracted from Baidu Baike,and 84 343 of them are linked to instances in DBpedia.We provide RDF dumps and SPARQL endpoint to access the established Chinese knowledge bases.The knowledge bases built using our approach can be used not only in Chinese linked data building,but also in many useful applications of large-scale knowledge bases,such as question-answering and semantic search. | Zhi-chun WANG Zhi-gang WANG Juan-zi LI Jeff Z.PAN | 2012 | Journal of Zhejiang University-Science C(Computers and Electronics)2012，13，4： | 11 |
| 3 | Effect of Berry Size and Sodium Hydroxide Pretreatment on the Drying Characteristics of Blueberries under Infrared Radiation Heating显示文摘 | J.Shi Z.Pan T.H.McHugh D.Wood Y.Zhu R.J.Avena‐Bustillos E.Hirschberg | 2008 | Journal of Food Science2008，，6： | 1 |
| 4 | Selective hepatic vascular exclusion versus Pringle manoeuvre in liver resection for tumours encroaching on major hepatic veins显示文摘 | J.Zhang E. C. H.Lai W.‐P.Zhou S.Fu Z.Pan Y.Yang W. Y.Lau M.‐C.Wu | 2012 | Br J Surg2012，，7： | 1 |
| 5 | Design for manufacturability and reliability in extreme-scaling VLSI显示文摘In the last five decades, the number of transistors on a chip has increased exponentially in accordance with the Moore's law, and the semiconductor industry has followed this law as long-term planning and targeting for research and development. However, as the transistor feature size is further shrunk to sub-14 nm nanometer regime, modern integrated circuit(IC) designs are challenged by exacerbated manufacturability and reliability issues. To overcome these grand challenges, full-chip modeling and physical design tools are imperative to achieve high manufacturability and reliability. In this paper, we will discuss some key process technology and VLSI design co-optimization issues in nanometer VLSI. | Bei YU Xiaoqing XU Subhendu ROY Yibo LIN Jiaojiao OU David Z.PAN | 2016 | Science China(Information Sciences)2016，59，6： | 1 |
| 6 | 查看详情显示文摘 | B.S.Ghanem K.J.Msayib N.B.McKeown K.D.M.Harris Z.Pan P.M.Budd A.Butler J.Selbie D.Book and A.Walton | | 0，，01： | 1 |
| 7 | Generative Learning in VLSI Design for Manufacturability: Current Status and Future Directions显示文摘With the continuous scaling of integrated circuit technologies,design for manufacturability(DFM)is becoming more critical,yet more challenging.Alongside,recent advances in machine learning have provided a new computing paradigm with promising applications in VLSI manufacturability.In particular,generative learning-regarded among the most interesting ideas in present-day machine learning-has demonstrated impressive capabilities in a wide range of applications.This paper surveys recent results of using generative learning in VLSI manufacturing modeling and optimization.Specifically,we examine the unique features of generative learning that have been leveraged to improve DFM efficiency in an unprecedented way;hence,paving the way to a new data-driven DFM approach.The state-of-the-art methods are presented,and challenges/opportunities are discussed. | Mohamed Baker Alawieh Yibo Lin Wei Ye David Z.Pan | 2019 | Journal of Microelectronic Manufacturing2019，2，4： | 0 |
| 8 | Challenges and opportunities toward fully automated analog layout design显示文摘Realizing the layouts of analog/mixed-signal(AMS)integrated circuits(ICs)is a complicated task due to the high design flexibility and sensitive circuit performance.Compared with the advancements of digital IC layout automation,analog IC layout design is still heavily manual,which leads to a more time-consuming and error-prone process.In recent years,significant progress has been made in automated analog layout design with emerging of several open-source frameworks.This paper firstly reviews the existing state-of-the art AMS layout synthesis frameworks with focus on the different approaches and their individual challenges.We then present recent research trends and opportunities in the field.Finally,we summaries the paper with open questions and future directions for fully-automating the analog IC layout. | Hao Chen Mingjie Liu Xiyuan Tang Keren Zhu Nan Sun David Z.Pan | 2020 | Journal of Semiconductors2020，41，11： | 0 |
| 9 | Local closed world reasoning:a personal view on current status and trends显示文摘1 Introduction In computer science and artificial intelligence,an ontology is a model of (some aspect of) the world that introduces the vocabulary of a particular domain,and specifies the meaning (semantics) of terms.Today's ontologies are based on the description logics (DLs).With the help of DL reasoners,implicit knowledge can be inferred from explicit knowledge in an ontology.Different from the relational database or logic programming,ontologies impose the open world assumption (OWA) instead of the closed world assumption (CWA).In CWA,a proposition will be regarded as false,if there is no evidence for it being true.While in OWA,a proposition will be regarded as false if and only if there is clear.evidence for it being false.These two assumptions seem to contradict with each other.However,in real world applications,it is usually desired to combine OWA and CWA.For example,a restaurant should not close the class ‘Vegetarian Customer’ because whether a new customer is vegetarian or not is unknown to the restaurant unless explicitly asserted by the customer.Nevertheless,it can close the class ‘Vegetarian Food Menu’ because it has the complete knowledge about its own menus.This raises the challenge of doing local closed world reasoning (LCWR) with ontologies.In the following sections of this extended abstract,we review the state of the art regarding this topic.We also propose several emerging topics in this research area and discuss possible technical directions. | Jeff Z.PAN | 2012 | Journal of Zhejiang University-Science C(Computers and Electronics)2012，13，4： | 0 |
| 10 | Editorial for Special Issue on Commonsense Knowledge and Reasoning:Representation,Acquisition and Applications显示文摘Commonsense knowledge is an important resource for humans to understand the meanings or semantics of the data.The ability to learn and own commonsense knowledge is one of the major gaps between humans and machines.Although recent research progresses about deep learning,like Transformer,pre-trained models,etc.,have made amazing breakthroughs in many fields,including computer version,natural language learning,etc.,letting the machines have rich commonsense knowledge and possess the reasoning ability is still difficult and under-resolved. | Kang Liu Yangqiu Song Jeff Z.Pan | 2024 | Machine Intelligence Research2024，21，2： | 0 |
| 11 | Finding Data Tractable Description Logics for Computing a Minimum Cost Diagnosis Based on ABox Decomposition显示文摘Ontology diagnosis, a well-known approach for handling inconsistencies in a description logic (DL) based ontology, computes a diagnosis of the ontology, i.e., a minimal subset of axioms in the ontology whose removal restores consistency. However, ontology diagnosis is computationally hard, especially computing a minimum cost diagnosis (MCD) which is a diagnosis such that the sum of the removal costs attached to its axioms is minimized. This paper addresses this problem by finding data tractable DLs for computing an MCD which allow computing an MCD in time polynomial in the size of the ABox of a given ontology. ABox decomposition is used to find a sufficient and necessary condition to identify data tractable DLs for computing an MCD under the unique name assumption (UNA) among all fragments of that are at least as expressive as without inverse roles. The most expressive, data tractable DL identified is without inverse roles or qualified existential restrictions. | 杜剑峰 漆桂林 Jeff Z.Pan | 2010 | Tsinghua Science and Technology2010，15，6： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员