PhD Candidate in Computer Science
Hi! I'm a PhD candidate in the Computer Science Department at Stanford University
where I work with Jure Leskovec and am part of SNAP, the InfoLab, and the Mobilize Center.
I enjoy analyzing large-scale traces of human activity to learn more about human behavior.
My research develops data science methods based on social network analysis, data mining, causal inference, and computational linguistics to harness the power of massive data to improve human health and well-being.
Most recently, I have worked on computational linguistic methods to analyze counseling and conversation strategies and data mining and social network analysis techniques to better understand exercise and health behaviors.
I have also conducted a study on Pokémon Go's impact on physical activity.
December 2016: Our paper on Pokémon Go and its impact on physical activity and public health together with Ryen White and Eric Horvitz at Microsoft Research just got published at the Journal of Medical Internet Research.
December 2016: I'm giving invited and plenary talks at the
GESIS Computational Social Science symposium,
the Max Planck Institute for Software Systems,
the German Research Center For Artificial Intelligence (DFKI),
the Friedrich-Alexander-Universität Erlangen-Nürnberg,
on leveraging data science and social networks to improve human health and well-being.
November 2016: Our paper on "Quantifying Dose Response Relationships Between Physical Activity and Health Using Propensity Scores" just got accepted to the NIPS Workshop on Machine Learning for Health.
October 2016: Our paper on Pokémon Go and its impact on physical activity has been covered by dozens of news outlets including
MIT Technology Review,
Reddit's front page.
It was also covered by several television news stations including the
Gotta catch ‘em all!
October 2016: I published a paper on Pokémon Go and its impact on physical activity and public health together with Ryen White and Eric Horvitz at Microsoft Research. We estimate that Pokémon Go added 144 billion steps to US physical activity within only 30 days.
August 2016: Our TACL paper on a large-scale analysis of crisis counseling conversations got featured in
May 2016: Kevin Clark and I published a blog post on the Stanford NLP blog highlighting some of the main results of our recent large-scale analysis of crisis counseling conversations.
May 2016: Our
on natural language processing for mental health analyzing a large annotated corpus of counseling conversations got accepted for publication at TACL.
More information on the project can be found here.
April 2016: I gave an invited talk at the "Wearable Devices & the 24-hour Activity Cycle" conference about "Consumer Mobile Tracking Devices as a Sensor for Physical Activity from Personal to Planetary Scale" organized by William Haskell,
Mary Rosenberger, and
March 2016: Hima Lakkaraju and I gave a lecture on submodular optimization in CS246 at Stanford. You can find the slides here. The featured paper on automatic timeline generation can be found here.
February 2016: I am honored to receive an Outstanding Reviewer award at WSDM 2016.
February 2016: I am excited to be working with Eric Horvitz and Ryen White at Microsoft Research this summer.
January 2016: My proposal on "Empowering Users Through Supportive Social Networks" was selected as a finalist for the Facebook fellowship.
Detection Bank: An Object Detection Based Video Representation for Multimedia Event Recognition
Tim Althoff, Hyun Oh Song, Trevor Darrell
ACM Multimedia 2012
Sparselet Models for Efficient Multiclass Object Detection
Hyun Oh Song, Stefan Zickler, Tim Althoff, Ross Girshick, Mario Fritz, Christopher Geyer, Pedro Felzenszwalb, Trevor Darrell
Proceedings of ECCV 2012
Don't Look Back: Post-hoc Category Detection via Sparse Reconstruction
Hyun Oh Song, Mario Fritz, Tim Althoff, Trevor Darrell
UC Berkeley, Tech. Rep. UCB/EECS-2012-16, Jan. 2012
Authorship Attribution in Multi-author Documents
Tim Althoff, Denny Britz, Zifei Shan
CS224U Natural Language Understanding (Chris Potts and Bill MacCartney) and
CS341 Project in Mining Massive Data Sets (Jure Leskovec, Jeff Ullman, Anand Rajaraman), Spring 2013/2014
Recommending CouchSurfers: Machine Learning for Social Travel
Ron Sun, Tobi Baumgartner, Tim Althoff, Sergey Karayev
CS281B Scalable Machine Learning (Alex Smola), Spring 2012
University of California, Berkeley
Iterative Learning: Leveraging the Computer as an On-Demand Expert Artist
Tim Althoff, Armin Samii
CS281A Statistical Learning Theory (Michael Jordan and Martin Wainwright) and
CS294-69 Image Manipulation and Computational Photography (Maneesh Agrawala), Fall 2011
University of California, Berkeley
Social Media Driven Concept Detection (survey)
Seminar on Collaborative Intelligence (Andreas Dengel), Summer 2011
University of Kaiserslautern
Jam-or-Fold: Eine spieltheoretische Betrachtung des Heads-Up Poker (English: Jam-or-Fold: A Game-theoretic Treatment of Heads-up Poker)
Freshman paper (Hausarbeit), September 2008
Eine mathematische Betrachtung des RSA-Verfahrens (English: A Mathematical Treatment of the RSA algorithm)
High School paper (Facharbeit), May 2006
email: [my last name] @stanford.edu
CV/references available upon request