Seminars


Big Data Colloquium Series

The Big Data Colloquium Series invites academic and industry leaders to share their insights on Big Data Science. The talks take place at the School of Computing and Information, University of Pittsburgh and are open to the public.

See upcoming events and past talks in the series.

Reading Seminar

Our weekly Research Seminar will discuss new techniques and key ideas in state-of-the-art data science research, covering topics including data mining, network science, and computational social science.

Unless otherwise noted, our seminars are held Friday afternoon at the IS Building. Please contact us if you have questions or are interested in joining us.

Data & Code Released

  • Bias Inflating with Ideology codepaper
  • Gun Narrative Tweets datasetpaper
  • StoCast Deep Generative Model codepaper
  • MimicProp: Learning to Incorporate Lexicon Knowledge into Distributed Word Representation codepaper
  • FairSight (Visual Analytics for Fairness in Decision Making) codepaper
  • CASTNet (Community-Attentive Spatio-Temporal Networks) codepaper
  • ActAttn (Interpretable Spatiotemporal Learning to Forecast Protest Activities) codepaper
  • DeepSphere (Anomaly Discovery in Dynamic Networks) codepaper
  • PairFac (Discriminant Tensor Factorization) codepaper
  • Measuring Collective Attention datasetpaper
  • MTHL (Multi-view Time-Series Hypersphere Learning) codepaper
  • Money and Politics dataset & codepaper
  • News data for studying media exposure dataset
  • Actionable Information During a Disaster (Self-organize Relief Efforts via #PorteOuverte) datasetpaper
  • Geo-tagged Tweets in Paris during Nov 2015 datasetpaper
  • Information Seeking in Academic Conferences datasetpaper
  • MetaFac (Relational Non-negative Tensor Decomposition)
    (This new release is compatible with MATLAB R2017a.) dataset & codepaper