News

About Me

I am a Ph.D. Candidate in the Computer Science Department at Stony Brook University, advised by Prof. Steven Skiena.

My research interest lies at the intersection of data mining, graph theory, and network science. I am particularly interested in local graph algorithms.
top

Research Interests

  • Data mining
  • Large-scale graph mining
  • Graph theory
  • Distributed algorithms
  • Natural language processing
  • Machine learning
top

Selected Publications

My complete publication list can be found here.
  • Statistically Significant Detection of Linguistic Change
    Vivek Kulkarni, Rami Al-Rfou, Bryan Perozzi, Steven Skiena
    24th International World Wide Web Conference (WWW '15)
    [pdf] [bibtex] [project page] [software]
    (full paper acceptance rate: 14.1%)
  • DeepWalk: Online Learning of Social Representations
    Bryan Perozzi, Rami Al-Rfou, Steven Skiena
    20th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '14)
    [pdf] [slides] [bibtex] [project page] [software]
    (full paper acceptance rate: 14.6%)
  • Focused Clustering and Outlier Detection in Large Attributed Graphs
    Bryan Perozzi, Leman Akoglu, Patricia Iglesias Sánchez, Emmanuel Müller
    20th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '14)
    [pdf] [slides] [bibtex] [project page] [software]
    (full paper acceptance rate: 14.6%)
top

Honors & Awards

  • 2015 - Catacosinos Fellowship for Excellence in Computer Science
    (Awarded to the top PhD students in the department)
  • 2014 - Junior Researcher Fellowship from the Institute for Advanced Computational Science (IACS) at Stony Brook University
top

Press & Social Media Reception

Various articles and comments on my work, from around the web:

News

Social Media
  • The New KDD -- highlights DeepWalk as an 'inspirational idea'.
top

Projects

  • DeepWalk uses deep learning techniques to learn representations of graphs for semi-supervised learning problems. [project page] [paper]
  • Focused Clustering examines user-oriented clustering and anomaly detection in attributed graphs. [project page] [paper]
  • Inducing Language Networks is a project where we investigate making meaningful language networks from distributed word representations. [project page] [paper]
  • Polyglot is a publicly available repository of word embeddings for over 100 languages. [project page] [paper]
top