前提:如果你是某个方向的,但是如果你没看过人家的论文,并且对其你的方向研究很不了解,这样的邮件写出来很没意思。会让人家觉得你虚假, 或者想不劳而获。那也就别怪人家不回复了。你再怎么语言得体,再怎么消极礼貌,都没用。
这是我网上收集的一点资料:

PART1 :注意点
  • 邮件主题(subject)不需要写自己的名字。名字是在邮件系统设置里设定好的,在发件时会替换显示实际地址,所以无需在主题里再写一遍。[补充:关键内容是要在主题里就写明白是否有推荐人,以及主要是什么事。很多教授只看主题。]
  • 在第一封邮件中称呼对方,两种正式的方法如下,其他的方式都是错误的。(例如 Prof.、Hello 、全名云云)
    Dear Dr. / Dear Professor * 是姓)
    一定要看清楚对方的简历, 确认其确实有博士学位, 才能用 Dr, 不然以后会很尴尬的 ……Professor 就无所谓, Assistant Professor 也是 Professor。 (可千万别写 Assistant Professor !=.=)*
  • 收到回信后,可以根据对方的落款,决定下一次如何称呼。如果对方落款只有名字 (firstname),那么下一次可以用名字称呼。如果对方落款是全名或是名字缩写,那么下一次仍然要用正式称呼。不过,在任何时候,用正式称呼都是保险的。
    如果是联系不认识的教授, 最好在见面之前始终用正式称呼; 如果是自己的导师或系里的经常会打交道的教授,那么可以尽快转到 first name ,可以拉近彼此的距离。此外,自己的身份和所处地方的文化也有很大影响。我曾经听一个美国朋友说,本科生称呼教授的 first name,是很不可想像的事情;也曾听另一个美国朋友说,她在中北部某校读硕士时,两年下来都一直称自己导师是 Dr.某,而且周围的同学也全是这个规矩。
  • 正文的第一句话必然是介绍自己。 这句话必然要遵循这个格式: My name is * (全名) ,and I am …例如:
    My name is Jeremy Bentham, and I am a graduate student at the Queen ’ s College, Oxford.
  • 正文的第二句话必然要解释自己是如何知道对方的。分三种情况:有人推荐;无人推荐;两人原本相识,但很久没有联系。

    这句话挺重要的, 因为它要在你和这个陌路教授之间建立第一道关联。 有人推荐当然最好,比如这个教授的同事某某某推荐你去找她,因为某某原因,那么把这个环节表述清楚即可。如果无人推荐 (更常见的情况) , 一定要说清楚原因, 例如刚刚看了人家的一篇新作云云 (你要确实看过才能这么说),不然人家可能就不往下读了。
  • 接下来便可以提出问题或请求了。这个部分遵循的法则叫做 “消极的礼貌 ”(negative politeness) 。 常用的策略有: 主动提及对方非常非常忙碌, 可能没有时间满足自己的请求 (也就是给对方一个台阶);多用 if 句;多用过去时态。[这是最最重要的两个原则之一。 我几乎每天都会在各类邮件里看到这个原则, 无论是学生对教授、教授对学生、教授对教授、还是学生对学生。后来我修常人方法学的课程,读到 Goffman、 Brown 和 Levinson , 真是拍案叫绝。 若想深入理解 “消极的礼貌 ”, 请读: Brown, P.,& Levinson, S. C. (1987). Politeness: Some Universals in Language Usage. Cambridge University Press.]
  • 避免使用 I want 和 I would like , 实在要用, 就用 I would love 。 最坏最坏, 也要用 I wanted 代替。[我从来没见过有人说 I want 或者 I would like ,即使是教授对学生。]
  • 多用 I was wondering if (一定要用 was,而不是 I wonder 或 I am wondering )。
    [打交道多了, 可以不必拘泥于此。 例如也可以说 I wondered... 甚至 shall we... 如果是自己的导师,或者工作往来很密切的教授,一天到晚说 “I was wondering if ,过分礼貌,其实也 ”挺奇怪的。]
  • 多用 would it be possible , 以便从语法上规避直接牵涉到 “人”, 这样可以减轻对方的压力。
    例如:(隐含的意思在括号里,但不可明写出来)
    Would it be possible (for you) to arrange an appointment (for me) some time?
    [很常用,我也经常收到教授用这个句法给我写的邮件。]
  • 避免使用 please,因为 please暗含了自己的期望,给对方压力,是不礼貌的表达。
    [这个规矩最宽松,但不熟的人还是不要用 please 为好。]
  • 提出请求后,一定要论证自己的请求是合理的。常用的方法是:展示自己已经做了大量的前期工作,实在是迫不得已,无路可走了,才来求人。或者,要表明自己对对方的工作了如指掌,而请求正是建立在这种了解之上的。[这是最最重要的两个原则之二。]
  • 结尾必然要感谢。最正式的表达是:
    Thank you, and I look forward to hearing from you.
    [后半句委婉地暗示着你希望她回复你的邮件。]
  • 落款之后的正式礼节问候是: Sincerely; Best wishes; Best regards。其他的都不甚正式。
    [熟了之后就可以多姿多彩了,例如可以写: Best, Cheers, Regards, See you soon, Nice weekend, 云云。或者学我们一个心理测量学家,写: Make sense. 不过,最常用也最百搭的一个或许是: Thanks. 尽管你可能根本没有提出请求,也可以用这个,谢什么呢? —— 谢谢对方花时间读了你的邮件。]

    PART2: 向外国学者索要论文,或者论文的源代码(网上的一个模板,觉得不错,但是觉得需要修改)
  • 向国外学者索取论文的模式

    Dear Mr./Mrs.: ________ (作者名)
      I am a graduate student of Xx University (大学名) in China. I major in "__Educational Technology______" (您的专业) . Recently, I found one of your articles, titled "__________" (文章名) in ---------( 杂志名). I found it may help me achieve my goals in this research field. This would make a really positive contribution to my work. I would like to be able to read the full text of this article. The abstract makes the article sound very interesting. I know there is usually a fee required to obtain the full article from -------- (杂志名) ; however, as a student, my only income is a small scholarship which is about U S $30.00 per month. I wonder if you would consider sending me the full text by Email. Perhaps you would consider this as an act of friendship between our two countries.Thank you for your kind consideration of this request.
    Sincerely: ___________ (您的名字)

    模板 1:

  • Wang's code works on a sample data set, but in my case it crashes. It's a data set with size 320x17000 and in lda-c format. Additionally, I couldn't see any instruction how to interpret the output files.
  • Teh's matlab package is a bit hard to follow. First of all, data format seems different than other formats. It accepts cell file, but I'm wondering whether I should give the document-word matrix incell structure.
  • The clustering task that HDP achieves is still not clear to me in a multi-corpora setting. Let's say I have Science and Nature corpora and HDP finds out 20 topics. Can anyone explain how we define clusters within each corpora? And is there a way to link cluster A in Science corpus to cluster A' in Nature corpus? IF they are sampled from the same topics, how can we manipulate the outputs of the code.
  • If there's any documentation or tutorial about coding, it would be also helpful.
    Thanks in advance

    我的一篇邮件:(当然,不是很好哈。但是基本表达清楚)

    Dear Professor Wu
    I am a first-year graduate student in XXX, China. My name is Tomas Cain. My research is focused on Graphic. And in this period, I am working in 3D reconstruction of vessel. And I have done some work about that.
    First, I have use the Volume Reconstruction Method, the Composite Method to implement on my DICOM data, which is provide by the hospital. A nd it ’ s effect is just ok. So, I take the next step .Then, I have implied the platform VMTK to implement the reconstruction of vessel. However, that platform ’ s Algorithm is based on the method using Simplex Meshes. And I cannot imply my own Algorithm in that platform. And, I have recently read your paper “Segmentation and reconstruction of vascular structures for 3D real- time simulation I am interesting in your paper, And the method proposed should have a better effect .
    I am wondering if you could kindly send me the source program codes or something necessary information about it. I promise they will be used only for research purposed. If it does, I will appreciate much for that.Thank you very much for your reading and I am looking forward to your early reply.
    Sincerely,
    Tomas Cain

    CONCLUSION :
最后修改:2021 年 10 月 08 日
如果觉得我的文章对你有用,请随意赞赏