How do you keep track of all the things you are working on? Do you have a physical notebook? Note taking application? Maybe just a text file?
Maybe a whiteboard is super helpful for keeping your key ideas around.
of course, all of us kind of started with a kind of proposal, how can we improve it, how can prevent from getting distracted from the main question? how can we know who worked on our question and maybe even solved that :(
what would you suggest in order to have some production-level code in NLP , which source would suggest, is it important at all? many thanks, Moha : ) Should we share it on GitHub?
Do you select a topic for your PhD which your supervisor is an expert in? What if your supervisor is not an expert in your topic, how do you deal with it? Are you looking for a different supervisor/university? If you continue with your supervisor, do you get feedback from other people in your research area? How to get in contact with people in your research area which are not in your department?
I'm currently in the situation that I'm working on a different topic than my supervisor and would like to discuss the situation with people who might are in a similar situation.
How do you develop NLP tools for languages without ready-made annotated datasets, maybe even without a standardised variety? What challenges do you encounter when creating your own datasets and there are no standards for annotation yet? What kind of approaches have you used or are considering to use?
I do not have any material prepared for this session, but in my PhD, I'm working on NLP for Low Saxon and, in addition, I'm involved in a machine translation project for Sami languages, so, in case there are other people here working on or interested in similar cases, I would like to discuss experiences, challenges you've encountered etc. with you.
As NLP PhD students, we work a lot with various tools and software. This includes general-purpose applications such as text processors, but also more specific utilities such as NLP libraries.
This session is for everyone who wants to discuss, suggest, or ask about these tools and to talk about their workflow with others. There is a very wide range of potential topics, including:
- Text processors and presentation software (Office, LaTeX);
- Note-taking and organization apps (e.g. OneNote);
- Operating systems (Windows vs. Linux vs. MacOS);
- Development environments (e.g. PyCharm);
- Programming languages and frameworks (R, Python, Pytorch, ...);
- Scientific and NLP libraries (Matplotlib, Huggingface, NLTK, ...).
Since there are a lot of fields in computational linguistics that are fast moving, I would like to get to know how you deal with regulations of your university regarding the publication of doctoral research. In my case, I write a monograph, so the communication of preliminary results to the community is not as easy as it might be with writing several research papers.
The other challenge I face is the publication of NLP results in interdisciplinary projects. How do you make your research part "visible" in such projects, especially when it comes to results that are intended for your doctoral studies?
This session is for everyone who needs or already has some ideas on best practice and who would like to share their own experience on publications and examination regulations.
Writing a paper and giving feedback to others is also an important part of a PhD. However, this requires a lot of training and is much more than a "magic" skill. In this session, we can talk about the writing process, the typical structure of a paper, helpful resources (online courses, etc.), writing tools, writing tandems, (constructive) feedback, and much more. I think this session can also be a good opportunity to network for writing tandems and find someone in your field to ask for feedback on your next draft.
A PhD is much more than writing papers. It can be a challenging journey with pitfalls, pressure, competition, rejections, failures, etc. that we may not talk about enough. However, I think it is important to emphasize that this is an essential part of a PhD and to create a platform to share ideas and support each other. In this session, we can share ideas, methods, and tips on how to manage stress during a PhD. What experiences have you had? How have you dealt with stressful situations? Who has helped you in these situations? What is the role of a valuable network of mentors and other doctoral students? What resources are helpful (university programs, online courses, etc.)? This session is for anyone who wants to share tips, ideas and experiences.
Digital Humanities (DH) is a relatively new field in which traditional questions of the humanities are attempted to be answered with computational methods. Doing a "NLP PhD" in this context can be challenging for different reasons: 1. There might be collaborations with researchers from the humanities for which a common language and/or goals need to be found first ("What are they talking about?", "Why do they not get what I am talking about?"), 2. the domain of the texts/material is very particular and standard NLP methods often do not work as well or need to be altered heavily, 3. or the question how much a dissertation should focus on NLP/CL or DH research questions (and what does that mean for my later career?)
I am doing a PhD in this field and would like to talk with you about your experiences, questions or ideas regarding anything that has to with doing a NLP-related PhD in the realm of DH.
The session is aimed at anyone doing such a PhD themselves, is thinking about doing one or shifting their current PhD more into this direction, or anyone simply interested in this topic. Depending on interest and attendees, we could also talk about doing a PhD in a generally interdisciplinary setting (regardless of the "other" discipline being from the humanities or not)
Of course, as computational linguists, we are both, somehow. But depending on your study background, you may rather feel like the one or the other. It can also happen that at some point, as in my case, you're switching from a rather linguistics-related background to a position that is closer to computer science (or vice versa). In this session, I would like to exchange thoughts about this "life between the disciplines". For example: Have you experienced differences between "Linguistics-NLP" and "Computer-Science-NLP" or do you feel it is the same? What do linguists know that computer scientists don't and vice versa? What would you like to learn (more) from the other discipline? This session is aimed at (PhD) students at any stage who would like to discuss where they see themselves in the spectrum of computational linguistics.
Attending a conference can be a lot of work, and it can sometimes be a bit scary, if you're going by yourself and you don't know anyone there. In this session, I will share my experiences from attending numerous IRL and virtual conferences, and give you tips on how to make the most out of your time at this type of event.
This session is aimed at conference first-timers as well as those who have attended conferences before, but are unsatisfied with how much they got out of those conferences for themselves. You are also invited to share your own tips and best practices with everyone.