Fred Morstatter

Excited to present our work "The Diffusion of Causal Language in Social Networks" as a plenary talk @IC2S2 this morning! In short, we found that causal language spreads more than non-causal language. Here is the link to our paper: ojs.aaai.org/index.php/ICWS…

Exciting news! Our new paper, Secret Keepers: The Impact of LLMs on Linguistic Markers of Personal Traits, is now available on Arxiv! Check it out! arxiv.org/abs/2404.00267 #NLProc #evaluation #LLM #psychology

Why LLMs are vulnerable to the 'butterfly effect' venturebeat.com/ai/why-llms-ar…

😍So excited for my 1st PNAS!!! 😍 We show that citation networks of men and women members of the National Academy of Sciences look so different that a machine learning algorithm can tell them apart! pnas.org/doi/10.1073/pn… with @YulinYuResearch @fredmorstatter & Jay Pujara

Buy this, @g1eb

New paper on Models, Markets, and the Forecasting of Elections papers.ssrn.com/sol3/papers.cf… (with Julie Seager, Emily Cai, Daniel Benjamin, and Fred Morstatter)

How has polarization, including anti-science attitudes, evolved since COVID-19 started? We explore this question with a new pre-print: arxiv.org/pdf/2011.08498… co-authors Ashwin Rao, @fredmorstatter, Minda Hu, @thatmchen, @emilio__ferrara and @KristinaLerman

@msalganik @KristinaLerman @psywisdom Do you know when they will stop accepting submissions?

@KristinaLerman They are still accepting submissions. You can find out more from @psywisdom.

Opportunity to participate in a mass collaboration. Can you forecast time series for 10 societal issues? Organized by Forecasting Collaborative: predictions.uwaterloo.ca

@psywisdom When will you stop taking submissions?

We still accept submissions!

@ColeRaid0311 @amylpeltekian lol obviously this isn't true. theintercept.com/2020/04/09/cor…

My first paper, "Aggressive, Repetitive, Intentional, Visible, and Imbalanced: Refining Representations for Cyberbullying Classification" (with @fredmorstatter and @YVigfusson), was accepted at #ICWSM 2020! Here is a link to the preprint: arxiv.org/pdf/2004.01820…

Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms window (old: 10ms). Not good for Science! #CSS #Twitter Thanks: @fredmorstatter @mehrnoom Background: Our 2018 paper on Twitter sampling: bit.ly/2JLaa5q

I am teaching a new course on Fairness in AI with @fredmorstatter - so new in fact, the syllabus is still evolving. If you have suggestions for papers we should include (especially for bias/fairness in networks) please send them to me. docs.google.com/document/d/16C…

Check out the @ISI_AI’s booth if you're at #ijcai19 . We have a number of exciting openings in ML, NLP, computer vision, and knowledge graphs. @USC_ISI @USCViterbi @IJCAIconf

Thank you ICWSM organizers for the wonderful company, interesting conference, and the abundance of delicious food! It’s been a 3-day long eating fest.

Honored!

Up next at the ISI Graduate Student Symposium! Social #bot behavior, repeatable testbed experiments and fraudulent #cryptocurrency manipulations on social media. #USC

New in @epj_ds Tampering with Twitter’s Sample API “due to the nature of Twitter’s sampling mechanism, it is possible to deliberately influence these samples, the extent and content of any topic” epjdatascience.springeropen.com/articles/10.11…