Publishing SCI papers from scratch: How does a novice publish the first computer-related SCI paper?

This article digs a big hole and will continue to be updated. The content is mainly divided into the following parts:

0. Tools (stationery)

1. Confirmation of topic selection

2. Read through field papers

3. Find a problem to solve

4. Motivation for the method

6 Experiments and data

7. Paper writing

8. Post

9. Reply to review comments (rebuttal)

10. Notification of employment

Currently, the author is writing and revising the first SCI paper, which has reached the seventh step. I will keep recording the experience during this period and provide some guidance for novices.

0 tools

If a worker wants to be good at his tools, he must first sharpen his tools (poor students have a lot of stationery).

The tools we use mainly include:

  1. Paper management tool Zotero
  2. Thesis typesetting tool texstudio (latex)
  3. Translation Tools (deepl.com)
  4. Polishing tools (write.youdao.com)
  5. anaconda (Python environment, drawing statistical graphs)
  6. Development ToolsVSCode
  7. Drawing tool draw.io (independent desktop version, drawing architecture diagram, schematic diagram, etc.)

 For draw.io, you must use the desktop version. The advantage is that it can directly export pdf files for easy use in latex. The vscode plugin can only export the svg version, and further conversion is required, which may be distorted during conversion. Using the native draw.io desktop version will save you these troubles.

That's right, we didn't use word. We will encounter a big problem of layout adjustment if we use Word for typesetting. As masters who will publish papers, we cannot use low-level tools such as word, which is too time-consuming. The difference between Latex and word is like matts and blank paper. In Word, you need to write the characters in a horizontal, flat and straight style on a white paper, but in matts, you only need to consider filling in the text, and you don't need to consider typesetting.

Important point: we use deepl.com to translate English to Chinese, not Chinese to English . The reason why many novices have problems is to translate Chinese into English. The correct way is to try to write the paper in English, and polish it properly after writing. Use deepl.com in English, Chinese and Chinese to check the accuracy of the expression. Using translation software to translate from Chinese to English is not a good way to write a thesis.

1 Confirmation of topic selection

Refer to the suggestions of those who have experienced it and your own interests, and choose a suitable direction to study. Pay attention to one point: remember not to change the direction after you are selected. Once you change the direction, it will take half a year. Your youth cannot afford to be wasted, and the doctorate is only four years. Before choosing, it is necessary to fully research the development of the field, the difficulty of publishing and the popularity. Emerging fields are given priority to dig holes and fill holes.

6 Experiments and data

After having an idea, it takes about 2 months to write the experimental code and run the data. This requires sufficient computing resources. The experiment is to run the data while changing. It may take about half a month to finish writing the first version of the program.

7 Paper writing

It takes about 4 months to write a dissertation. Including first draft, revision, language polishing and polishing.

7.10 Overall Proofreading

The paper was published and revised 10 times. After writing, it is necessary to polish the text. The method is to print it into a paper version, and use a highlighter to draw sentences that are not smooth or natural while reading. until smooth. If possible, you can find someone whose native language is English to help you read it, find unnatural expressions and modify them.

8 posts

Posted on 1.21st

Received the first round of approval notice on 2.23

Guess you like

Origin blog.csdn.net/znsoft/article/details/128623935