Tips for writing a research paper


Spread the love

The following recommendations for conducting research work (R & D) are addressed to students and can be useful at all stages of training, from the selection of a supervisor to the preparation of the final qualifying work.

Individual recommendations may be specific to the field of machine learning (for example, the balance of theory and practice, mathematics and heuristics), and other areas should be transferred with caution.

The main steps

Working with a supervisor usually begins with solving several test problems and / or reading several articles on the topic of a future R & D, possibly in English. Then you will be given the main task, which in the future should turn into the theme of final work.

Having received the next task, do not hesitate to ask for additional explanations. Much worse, if you are buried, you will perish for a long time, without having done anything.

Another common mistake is to postpone research paper until later. Usually the manager expects that your work will join the general research and expect certain results by certain deadlines. If you cope with the first task quickly, you get a complication, then the next one, and by the end of the study you will have tangible results. If a student remembers about research at the end of the semester (year, last year) and begins to solve the initially set simple task, then the work is too simple and unnecessary. It’s almost impossible to do something worthwhile in a couple of weeks. It is not surprising if such coursework or final work will be rated no higher than three.

Research paper need to deal with constantly. Good ideas emerge as a result of repeated joint discussions, and not immediately. It takes time to figure out the reasons for the failure of the first experiments, to come up with the best algorithm, to understand something and prove its properties. Therefore, a good term paper can objectively be done only in a couple of semesters.

Your manager has the right to be busy, not to find time to read the material you have sent for a whole week (two, three, …), once again postpone the meeting another week later, and provide other signs of inattention. This is normal. Spend the resulting time on it to independently understand what to do next or do self-education. No one will give a student a topic on which there is not a single publication in the world. Do not forget about Google search, Google Scholar, arXiv, CiteSeer, Sci-Hub.