《如何撰写与发表高水平学术论文.ppt》由会员分享，可在线阅读，更多相关《如何撰写与发表高水平学术论文.ppt（54页珍藏版）》请在三一办公上搜索。
1、如何撰写与发表高水平的学术论文,How to Write and Publish High Quality Scientific Paper,贺昱曜 教授,2007年11月13日,一、学术论文的特征与意义-Characteristic and Significance of Scientific Paper二、标题的创作-Creating the title三、摘要的写作-Writing the abstract四、引言的撰写 论文写作目的与创新点的导引-Writing the Introduction 五、正文与结果的撰写要点-Writing the Main Body and Result
2、s of a Paper六、结论的写作技巧-Writing the Conclusion七、致谢-Stating the Acknowledgements 八、参考文献-论文前沿性的有力支撑-References Citing 九、投稿从主编和评审人的角度考察论文-Manuscript Submitting,主要包含以下九个方面,一、学术论文的特征与意义-Characteristic and Uniform Requirements for Scientific Paper Science（科学）Knowledge covering general truths or the operatio
3、n of general law.Natural Science（自然科学）A system or method reconciling practical ends with scientific law.Applied Science（应用科学）Search for practical uses of scientific knowledge.Study（研究）Critical and exhaustive investigation or experimentation aimed at the discovery of new facts and their correct inter
4、pretation,or the revision of accepted theories or laws in the light of new facts,or practical application of such new or revised theories or laws.Research Article（学术论文）:Paper and Dissertation or Master Thesis for the degree of Ph.D or M.E,1.1 为什么要发表论文与学术论文的意义：硕士、博士毕业；技术人员，教师评职称；申报学科点，重点学科，重点实验室；申报项目
5、，申报奖励；一个学校、学院的学术声誉；一个人的学术成就须用论文的数量与水平来衡量，等。,1.2 Definition of Scientific paper Scientific paper should include creative points-new ideas.The results and conclusions can be verified by others-经得起检验的结果和结论.New ideas New concept TheoryDeduction Equation Method(approach)Algorithm Scheme Strategy New Topo
6、logy Circuit topology,System architecture New Hardware Software 只有包含创新点的文章才能称作为论文。,你的题目有没有学术价值？有没有新东西？-新观点、新模型：过去没有人这样看问题；-新方法：过去没有人这样分析、解决问题；-新结论：过去没有人得到这样的结果。,1.3 Journal,Transaction,Magazine,Letters Journal of Northwestern Polytechnical University Journal of Changan University IEEE Transaction on
7、 Automatic Control IEEE Transaction on Neural Networks Power Electronics Communications Computer;Circuits and Systems I Vision,Image and Signal Processing.IEEE Acoustics,Speech,and Signal Processing Newsletter IEEE Journal of Oceanic Engineering IEEE Communications Magazine IEEE Communications Lette
8、r,1.4 Importance of Scientific Paper Writing Three functions of Scientific Paper:(1)Record and store information;(2)Communicate with others;(3)Help to creativeness.a.Accumulate the knowledge;b.Measure and become known by their publications;c.Refine the research results;国外没有我们国内这么多的评奖，评价一个人的成就主要看高水平的
9、论文。例如杨振宁的诺贝尔奖 The goal of scientific research is publicationScientists,starting as graduate students，are measured primarily not by their dexterity in laboratory manipulations,not by their innate knowledge of either broad or narrow scientific subjects,and certainly not by their wit or charm;they are
10、measured,and become known(or remain unknown)by their publications.,1.5 Characteristics and Requirements for Scientific Paper Writing（1）Creativity 创新 Original manuscript which has not been published anywhere else.科学论文所介绍的研究成果应该是前所未有的，即有所发现、有所创造。没 有新的观点、见解、结果和结论，就不成其为科学论文。创新程度可大，可小，总要对科技知识宝库有所贡献。“首先提出
11、”、“首次 发现”具有重大意义的研究毕竟不多。就某一方面而言有新意、有发展，亦应属于 创新。有些课题是引进消化国内外先进技术，只要对丰富理论，推动技术进步有 意义，也应视为有一定程度的创新。（2）Clarity 简洁（3）Clearness 明了（4）Accuracy 精确（5）Standardization 标准 Understandable,1.6 Classification of Scientific Papers(1)Theoretical analysis Assumption,Principle,Theorem,Lemma,Law and proof.(2)Application
12、 study Measurement,test,data processing.(3)Mathematical modeling and computer simulation(4)Review or survey,overview,1.7 Organization of a scientific paper,Title Authors Abstract Keywords Introduction Material and Method Results Conclusion and Discussion Acknowledgement References Appendixes,二、标题的创作
13、-Creating the title Procedure for writing a scientific paper(1)Body Results Conclusion(2)Introduction References Abstract Title,2.1 Functions and importance of the title Title for the paper is like the face for some beautiful girl.Outline the paper.Summarize research contents(new ideas)and results a
14、nd be convenient to index 阅读习惯：先看标题，根据标题来考虑是否看摘要。标题的功能是概括全文并便于检索索引。TClifford Allbutt的话，不仅精辟地说明标题的重要性，也是写标题的指导思想：First Impressions are strong impressions；a title ought therefore to be well studied，and to give，so far as its limits permit，a definite and concise indication of what is to come 第一印象是最重要的印象
15、；因此标题应当仔细研究，在允许情况下，尽可能确切、简明地反映所要讲的内容。,2.2 Requirements for the title(1)Expressing the practical contents and characteristic of the paper.(2)Clarity and accuracy 10-12 words(3)Confirming with the common usage of title writing.Example 1.A singular perturbation finite element procedure for the analysis
16、 of edge effects in shell structures 奇异摄动有限元方法分析壳结构的边界效应 What method for what problem.Example 2.Shape design sensitivity analysis of a two dimensional heat transfer system using the boundary element method 用边界元方法分析二维传热系统的形状设计的灵敏度（what）problem analysis using what method.Example 3.Numerical analysis o
17、f Large Deformation Rigid Plastic Problems in Dynamic Range by Quadratic Extremum Principles.动力学范围内的大变形刚朔性问题的二次极值原理数值分析 一篇论文只能有一个主题，绝对不能有两个以上的主题。,三、摘要的写作-Writing the abstract How to scan a paper?Title Abstract Conclusion Body How to index a paper?Title Abstract Author Journal Affiliation Study Criti
18、cal and exhaustive investigation or experimentation aimed at the discovery of new facts and their correct interpretation,or the revision of accepted theories or laws in the light of new facts,or practical application of such new or revised theories or laws.,3.1 Definition of the Abstract An Abstract
19、 should be viewed as a mini-versionof the paper.The Abstract should provide a brief summary of each of the main section of the paper.An abstract can be defined as a summary of the information in a document.,3.2 Importance of the Abstract 英国前首相丘吉尔说过：Please be good enough to put your conclusions and r
20、ecommendations on one sheet of paper at the very beginning of your report,so that I can even consider reading it.,1.Academic communication 2.Indexed by EI,SCI or ISTP 3.If you submit a paper,the editor may make a decision by your abstract.,Because the abstract precedes the paper itself,and because t
21、he reviewers and editors like a bit of the orientation，the Abstract is universally the first part of the manuscript read during the review processTherefore，it is of fundamental importance that the Abstract should be written clearly and simply Very often，the reviewer may be perilously close to a fina
22、l judgment of your manuscript after reading the Abstract alone.It is only logical that the reviewer will often reach a preliminary conclusion，and the conclusion is likely to be a correct one Usually，a good Abstract is followed by a good paper；a poor Abstract is a harbinger of woes to come,3.3 Types
23、of the Abstract Two types of the abstract通常分为信息型(informative)或提供资料的摘要和叙述型(descriptive)摘要两种。1.Descriptive abstract 描述研究的范围和概况，不介绍有关的方法和结论。,Tell readers what information the paper contains;Include the purpose,methods,and scope of the paper;Do not provide results,conclusions,or recommendations;Are alwa
24、ys very short,usually under 100 words;Introduce the subject to readers,who must then read the paper to find out the authors results,conclusions,or recommendations.,Review or survey paper.,2.Informative abstract 信息型摘要或称为提供资料的摘要，不仅介绍所研究的课题，而且提供详细的研究结果。它是论文全文的浓缩，往往可以代替阅读全文。,Communicate specific informa
25、tion from the paper;Include the purpose,methods,and scope of the paper;Provide the papers results,conclusions,and recommendations;Are short from a paragraph to a page;Allow readers to decide whether they want to read the paper.,3.4 Uniform requirements for the contents of the abstract 摘要的必要内容是：研究工作的
26、背景，目的，方法，结果和结论，重点是结果和结论。必须强调：由于科学论文的根本特征在于创新，摘要的内容必须体现这个特征，一定要表明本文的特殊贡献。,Day书中也是归纳为4点：(Day：1994，30)The Abstract should(1)state the principal objective and scope of the investigation，(2)describe the methods employed，(3)summarize the results and(4)state the principal conclusionsThe importance of the c
27、onclusion is indicated by the fact that they are often given three times：once in the abstract，again in the introduction，and again(in more detailed probably)in the DiscussionThe Abstract should never give any information or conclusion that is not stated in the paper。摘要应该(1)说明主要目的及研究范围，(2)描述所用的方法，(3)概
28、括结果，(4)说明主要结论。结论的重要性可以用如下事实表明：通常在文章中三次出现。第一次在摘要中，再次在引言中，最后还出现在讨论或结论中(可能更为详细)。现在有些杂志明确地提出“结构化摘要Structure abstract”。整篇摘要分为四段，其标题为：Background，Methods，Results，Conclusion。不言而喻，这种形式很有优越性。,Background&Object Method Results Conclusion 1.Background&Object What are the current states of your research field?Wha
29、t problems exist in that field-open problems?What special problem do you want to solve?2.Method What method did you use to solve the problem?What formula did you obtain?What is new in your paper,Figure,Circuit,equation or approach?3.Results Simulation results Experimental results Illustrative exampl
30、e Numerical results,Figures,Tables.Comparative with other results 4.Conclusion Have you solved the problem?Is your new method validity?What other problem can your new method apply for?What are the advantages of your method?,Chaotic Neural Networks and Their Applications Yuyao He Lipo Wang College of
31、 Marine Engineering Shool of Electrical and Electronic Engineering Northwestern Polytechnical University Nanyang Technological University Xian,Shaanxi 710072,P.R.China Block S2,Nanyang Avenue,Singapore 639798Abstract-Many difficult combinatorial optimization problems arising from science and technol
32、ogy are often difficult to solve exactly.Hence a great number of approximate algorithms for solving combinatorial optimization problems have been developed 10,15.Hopfield and Tank applied the continuous-time,continuous-output Hopfield neural network(CTCO-HNN)to TSP,thereby initiating a new approach
33、to optimization problems.But Hopfield neural network is often trapped in local minima because of its gradient descent property.A number of modifications have been done on Hopfield neural network for escaping from local minima.As so far,incorporating chaos into the Hopfield neural network has been pr
34、oved to be successful approach to improve the convergent property of the HNNs.In this paper,we first review three chaotic neural network models,and then propose a novel approach to chaotic simulated annealing.Second,we apply all of them to 10 city TSP,respectively.The time evolutions of energy funct
35、ions and outputs of neurons for each model are given.The features and effectiveness of four methods are discussed and evaluated according to the simulation results.We conclude that proposed neural network with simulated annealing has more powerful ability to obtain global minima than any other chaot
36、ic neural network model when applied to difficult combinatorial optimization problems.Keywords:Neural Networks,Transient Chaos,Simulated Annealing,Combinatorial Optimization Problems,TSP This paper was indexed by EI 2003年刊 NO.114832,3.5 Tips for preparing the Abstract(摘要的写作要点)1.常用动词 Study Analyze Pr
37、esent Propose Develop Investigate Prove Derive Obtain Deal with Discuss Introduce Provide Give Point out Review Conclude Complete Accomplish Perform Invent Examine Discover,四、引言的撰写 论文写作目的与创新点的导引-Writing the Introduction 4.1 Importance of the Introduction How to guide the readers to get what problem
38、you have solved,what is your new idea,and what new results you have gotten.引言-主要任务是说明本文的写作理由，本文的工作与前人工作的关系，以及帮助读者了解正文的背景、事实和条件。正文-着眼与对所提出的问题的描述，分析，论证。摘要-重点是介绍本文的主要贡献，而没有更多的篇幅来系统的介绍本文的背景，与前人工作的联系与区别。How to introduce clue to readers from general problem to a special problem?What is new idea,new facts,
39、and new theories or laws?引言的作用是提供足够的研究背景和内容，引领读者快速进入论文主题，领悟作者研究工作的构想与设计思路，正确理解和评价论文中研究的结果。引言通常包括以下几方面信息：为什么写这篇论文，要解决什么问题；与课题相关的历史回顾，本课题与前人工作的关系，在学科领域中所占的地位及课题的意义和价值；本研究所涉及的界限、规模和范围；理论依据和实验设备基础；预期目标。,4.2 Contents of the Introduction 问题的提出，研究范围与目的，主要研究方法。1Background of the problem and chief contributi
40、on of others问题的背景及他人的贡献 首先是从问题的工程背景与历史背景开始，介绍并评论国内外其他学者对该问题或现象曾经发表的相关研究，即文献回顾。选择引用密切相关的文献简述，问题的理论或实用意义及他人在该领域的工作，前人工作的优缺点及存在的问题，系统的列举有关参考文献。2Purpose,scope,and contribution of the study 研究范围、目的及贡献 在广泛的背景下，先从普遍问题开始，然后逐步缩小讨论问题的范围，最后将重点落在与论文所探讨的问题密切相关的主题上。指出仍然有某个问题或现象值得进一步研究，即将本课题的任务具体化，点明本文研究的意义。说明所要解决
41、的具体问题，完成的工作，也就是说在已有工作的基础上，有什么贡献，或者明确交代有那些创新。3Methods of attack 主要研究方法 分为三类：,引用前人的方法，但解决了新问题；对已有方法的修正，或改进；提出新的方法，理论或定律。,期刊所关心的首先是论文的贡献。其中有什么创新，有什么使用价值或应用前景。这些内容是引言的重点。实例(Illustrative examples)I.INTRODUCTION Combinatorial optimization problems arising from science and technology can be described as a
42、search for the best solutions.Because a number of interesting combinatorial optimization problems are difficult to find exact solutions due to their NP hardness,many researchers have been paying much attention to approximate or heuristic algorithms 1,2.Since Hopfield and Tank applied Hopfield Neural
43、 Network(HNN)3 to solving the traveling salesman problem,the HNNs have been recognized as useful tools for combinatorial optimization problems 4,5.However,using HNNs to solve the TSP suffers from several shortcomings.First,the network can often be trapped at a local minimum on the complex energy ter
44、rain.Second,an HNN may not converge to a valid tour.Third,an HNN sometimes does not converge at all within a prescribed number of iteration 6.Chaotic neural networks have recently received attention due to the potential capability for information processing 7,8.In a number of methods proposed to pre
45、vent the HNN from being trapped at local minima,chaotic neural networks have been shown to be powerful approaches.Based on chaotic properties of biological neurons 9,Chen and Aihara recently proposed chaotic simulated annealing to illustrate the features and effectiveness of a transiently chaotic ne
46、ural network(TCNN)in solving optimization problems 10.By adding a negative self-coupling to a Hopfield-like neural network and gradually removing it,they used the transient chaos for searching and self-organizing.Compared to other neural network methods 11,12,the Chen-Aihara approach significantly i
47、ncreased the probability of finding near-optimal solutions.However,the Chen-Aihara method requires that a number of network parameters must be subtly adjusted so as to guarantee the convergence of the TCNN and its minimization of the energy upon the removal of the transient chaos 13.,Wang and Smith
48、presented an alternate approach to chaotic simulated annealing 14.Starting from the Euler approximation of the continuous-time,continuous-output Hopfield Neural Network(CTCO-HNN)with a large initial time step,where the dynamics is chaotic,and then gradually reducing the time step,the system is guara
49、nteed to converge and to minimize the HNN energy function.Since it is in the limit of,the network approaches the CTCO-HNN.This approach eliminates the need for difficult choices of any other system parameter.Hayakawa et al 15 proposed another approach for TSP by adding chaotic noise to the discretiz
50、ed CTCO-HNN.Since the system with chaotic noise does not converge,a statistical average over a period of time is used to determine the state of a neuron.Because this technique has hill-climbing features,the rate of reaching global optimization is lower.In this paper the chaotic noise is added to eac
