Anne Dirkson, Suzan Verberne, Gerard van Oortmerssen, Hans Gelderblom, Wessel Kraaij. How do others cope? Extracting coping strategies for adverse drug events from social media. Journal of Biomedical Informatics. 2022 Oct 26:104228. doi: 10.1016/j.jbi.2022.104228 PMID: 36309197. https://www.sciencedirect.com/science/article/pii/S1532046422002337
Her thesis is not online yet but if you are interested in the topic you could have a look at these two papers from her thesis:
Anne Dirkson, Suzan Verberne, Wessel Kraaij, Gerard van Oortmerssen and Hans Gelderblom. Automated gathering of real-world data from online patient forums can complement pharmacovigilance for rare cancers. Nature Scientific Reports 12, 10317 (2022). https://doi.org/10.1038/s41598-022-13894-8
Next Tuesday is the PhD defence of Anne Dirkson:
“Knowledge Discovery from Patient Forums. Gaining Novel Medical Insights from Patient Experiences”
I am proud of her and her work; she did a lot of interesting work and got relevant results 🙂
My review of the textbook "Pretrained Transformers for Text Ranking: BERT and Beyond" by Jimmy Lin, Rodrigo Nogueira, and Andrew Yates has been published in Computational Linguistics journal:
@suzan @mdekstrand To add some of my own talking points:
- highlight a few ways recommendation is *not* matrix completion (explicit feedback, data not missing at random, ...) but closer to decision theory, and how RL comes into the picture as a result
- real-world systems consist of different stages (retrieval, re-ranking, and others) for "engineering" reasons that can be fascinating in their own right
LIACS is hiring 5 Assistant/Associate/Full Professors in Artificial Intelligence and Computer Science.
Come join us in Leiden 🙂
@suzan I do a 90 min lecture in our decision support class with (roughly) this outline: definition (e.g, ranked personalized list of items taking the context, situation, and information need into consideration), purpose (overchoice, information overload) (not forgetting the economical incentive),
basic CF (user- and item-based) and hybrid methods, datasets (user-item matrix, what data points do we use) (combined with a walkthrough of kNN with graphical examples), (1/n)
@suzan distinct recommendation task families (big 4 I’d pick: front page, session, related-product, streaming). Role of context and collaborative signals in RecSys, how CF works in principle. Modeling preference through user embeddings, and how users and items embed in the same space.
Avishek Anand and his students wrote a survey on EXplainable Information Retrieval, EXIR.
If I want to include one lecture (2x 45 minutes) about #RecSys in my Information Retrieval course, what topics should I cover?
"Dutch universities published 82% of their peer-reviewed scientific publications #OpenAccess in 2021. This is an increase of nine percentage points compared to 2020."
I am hiring a PhD student in the project LESSEN (https://lessen-project.nl) on conversational agents. It is a 4-year, fully paid position
You will be embedded in Text Mining & Retrieval Leiden at LIACS, Leiden University, supervised by me, and co-supervised by @fhasibi
Norbert Fuhr's recommendations for gaining scientific knowledge from experiments:
1. Do not use MRR or MAP;
2. Instead of relative improvements, regard the effect size!
3. For multiple significance tests, use a correction, such as Bonferoni or Tukey's HSD (NB comparing only to the 2nd best method does not help!)
4. There are no significant improvements for re-usable test collections! (hypotheses have to be formulated before the work)
Text Mining & Retrieval, Leiden University
The "unofficial" Information Retrieval Mastodon Instance.
Goal: Make idf.social a viable and valuable social space for anyone working in Information Retrieval and related scientific research.
Everyone welcome but expect some level of geekiness on the instance and federated timelines.