#introductions Hey everyone. I am a PhD Student in Computational Linguistics. My main research interest is computational linguistics for under resourced languages. I am particularly interested in computational linguistics for Bantu languages because of the extremely complex morphology that they bring to the table.
I am also doing research in sentiment analysis on stance detection and irony/sarcasm detection. Most of my research is on machine learning approaches.
@ksteimel Hi welcome!
@bgcarlisle Hey thanks!
Sorry, quick question. Do people talk about non-academic things as well on here or is that generally frowned down on? Nothing inappropriate just cat pictures.
@ksteimel Generally speaking, you want to post "on-topic" stuff as Public and "off-topic" as Unlisted
And there's CW's which are appropriate for politics, FB/Twitter mentions, lewd/nsfw stuff, spoilers for TV shows, or even hiding the punch line of a joke!
Guessing: this is a complex time for your research?
@katebowles Hi! Nice to meet you! Yes I'm involved with a lot of different projects as well as my own independent research. Trying to get involved with everything I can before I have to go into dissertation mode. I need to find two disparate topics for qualifiying exams this summer.
@ksteimel I had no idea computational linguistics was tackling irony/sarcasm. That's awesome. I had a brief glimpse at computational linguistics when I worked for a project called STEDT in the early 1990s at UC Berkeley. Neat stuff.
Oh cool! Yeah part of how I'm approaching it is by trying to detect shifts in the sentiment of the words used. For example, 'A beautiful spring day \#brr' would have a shift in sentiment from positive at the beginning to negative at the end.
There's also certain words that tend to indicate sarcasm like 'just'.
If you don't mind me asking, what was STEDT working on?
@ksteimel STEDT was the Sino-Tibetan Etymological Dictionary and Thesaurus. We were compiling a database of Sino-Tibetan languages which allowed comparison across them. I walked in the door knowing zero about linguistics which meant I was learning at every turn. It was really fun.
@ksteimel Great to meet you. Do you sit more on the computational or the linguistic side of things?
I am usually on information retrieval / computational text understanding side. Usually English, but I also work with "crosslingual" folks.
I'm more on the linguistics side of things but I do have a huge interest in parallelizing NLP algorithms.
Oh awesome! It's great to meet someone else in my field.
What do you do for text understanding?
I had a class last semester where we took dependency parses and extracted semantic relations to put in a graph dB but I don't know if this is actually text understanding or really what text understanding entails
@ksteimel This is an "accepted approach" but I disagree.
I run a shared task on Complex Answer Retrieval. Given a search query, generate a Wikipedia page that tells you everything about it. http//trec-car.cs.unh.edu
(Sorry, website needs updating)
The manual evaluation step must take a while. This looks like a very cool albeit difficult shared task.
Thank you for showing me this!
@ksteimel Well, you can use all of WIkipedia to train and do a first step evaluation
NOTICE: Registration on scholar.social is open to anyone who is willing to abide by our Community Standards. Email scholar dot social at protonmail dot com if you want an invite!
Scholar Social is a microblogging platform for researchers, grad students, librarians, archivists, undergrads, academically inclined high schoolers, educators of all levels, journal editors, research assistants, professors, administrators—anyone involved in academia who is willing to engage with others respectfully.
We strive to be a safe space for queer people and other minorities in academia, recognizing that there can only be academic freedom where the existence and validity of interlocutors' identities is taken as axiomatic.
"An academic microblog that you can be proud to put on the last slide of a presentation at a conference"
(Participation is, of course, optional)
Scholar Social features a monthly "official" journal club, in which we try to read and comment on a paper of interest.
Any user of Scholar Social can suggest an article by sending the DOI by direct message to @firstname.lastname@example.org and one will be chosen by random lottery on the last day of the month. We ask that you only submit articles that are from *outside* your own field of study to try to ensure that the papers we read are accessible and interesting to non-experts.