As the computational social scientist (aka data scientist) at Humanyze, I design new algorithms and metrics to reveal how works get done in teams and organizations, using face-to-face and digital communication metadata collected from wearable sensors, IoT devices, as well as from data integration APIs of communication and collaboration tools (such as calendar, Skype, email, etc.). The theories and methods on which my algorithms rely are primarily from the fields of organizational behavior (management science) and social network analysis (network science). I not only design and develop algorithms by integrating organizational theories with new kinds of digital and sensor data but I also work on advancing the science of organizational behavior by conducting experiments and writing scientific papers.


About. I strive for understanding complex social and behavioral phenomena. To this end, I’ve been studying towards my Ph.D. in computational social science (CSS) in the Department of Computational & Data Sciences (CDS) at George Mason University (GMU). My background is in computer science (CS); actually, I hold two master’s degrees in CS, one with a focus on data mining and the other on internet measurements. For about five years, I have worked in the Machine Learning and Inference lab (MLI) within the Center for Discovery Science and Health Informatics (DSHI). Then for a year (before joining Humanyze) I worked at the Center for Social Complexity within Krasnow Institute on population synthesis for an agent-based model of Characterizing Response to a WMD Event in a Mega-city, supported by DTRA.

More About Me…

On this personal site you can find more about my background and current research, and you can read my blogs in English and Turkish. If you are so inclined, you can follow me on Twitter and gitHub.

You can reach me at '@'.join(['toz', 'gmu.edu'])
Regarding my professional activities, you can find my professional resume and past research projects


Some of my past work:

  • What Happens If a Nuclear Bomb Goes Off in Manhattan? [Notebook]
  • Attribution of Blame and Responsibility in #FlintWaterCrisis [Notebook]
  • Politicians Busted while Agenda-setting on Social Media [Blog post w/ repo]
  • Twlets: Twitter→Excel, Chrome Web Browser Extension [web app, code]
  • AirBnB++: Search Listings by Reputation and Description [Notebook]
  • GeoPopularity of Politicians [Notebook]
  • Doomsayers or Pollyannas? Gatekeepers on Twitter [Paper, Notebook]
  • Zotero-picker plugin for Atom text editor [Plugin page w/ repo]


I have taken so many classes. Some of them were good, and others were great. Here is a list of my favorite classes:

George Mason University (CSS, Ph.D.)

  • Computational Analysis of Social Complexity (CSS 610)
  • Spatial Agent-based Models of Human-Environment Interactions (CSS 645)
  • Complexity Theory in Social Sciences (CSS 625)
  • Geosocial Media (CSS 739)
  • Geographic Information System (GGS 553)
  • Geospatial Intelligence (GGS 684)
  • Web-based GIS (GGS 692)

George Mason University (CS, MS)

  • Big Data
  • Data Mining on Multimedia Data
  • Data Mining
  • Algorithm Analysis
  • Social Networks
  • Artificial Intelligence
  • Semantic Web & Knowledge Engineering
  • OO Software Specification & Construction
  • Software Modeling

University of Nevada-Reno (CS, MS)

  • Complex Networks
  • Computer Networks
  • Internet Protocols
  • Genetic Algorithms
  • Combinatorics & Graph Theory

MOOC (Online learning)

I think MOOC is one of the greatest blessings we have just started enjoying. I have learned a lot especially from some Coursera courses such as Data Analysis and Statistical Inference and Model Thinking, and looking forward to learning more.

I also enjoy online hackathons, in particular, I used to actively participate Kaggle competitions, and once was a top percentile data scientist [it appears that their ranking algorithm takes active participation into account seriously].