Abhinandan Das

Abhinandan Das

Dr Abhinandan Das received his Ph.D. in Computer Science from Cornell University in 2005. He has authored several research publications in leading conferences and journals on databases, data mining and data stream processing (including a best paper nomination at the World Wide Web conference). His published work has been deployed in production systems in companies such as Google, Uber, and Microsoft, among others. He holds 20+ issued and filed patents in diverse areas such as real-time Recommendation Systems, Collaborative Filtering, Personalization, Data Mining, Related Search, Query Suggestions, Web Search Ranking, Autocomplete Systems, Query Rewriting and Spell Correction, Text Input Likelihood Prediction, Name Disambiguation, Distributed Group Membership Protocols and Distributed Systems Monitoring, among others.
Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, descending
  • Year
  • Year, descending
    Google News Personalization: Scalable Online Collaborative Filtering
    Mayur Datar
    Ashutosh Garg
    Shyam Rajaram
    Proceedings of WWW 2007, pp. 271-280
    Preview
    Semantic Approximation of Data Stream Joins
    Johannes Gehrke
    Mirek Riedewald
    IEEE Trans. Knowl. Data Eng., 17(2005), pp. 44-59
    Approximation Techniques for Spatial Data
    Johannes Gehrke
    Mirek Riedewald
    SIGMOD Conference(2004), pp. 695-706
    Distributed Set Expression Cardinality Estimation
    Sumit Ganguly
    Minos N. Garofalakis
    Rajeev Rastogi
    VLDB(2004), pp. 312-323
    Efficient Approximation of Correlated Sums on Data Streams
    Rohit Ananthakrishna
    Johannes Gehrke
    S. Muthukrishnan
    Divesh Srivastava
    IEEE Trans. Knowl. Data Eng., 15(2003), pp. 569-572
    Approximate Join Processing Over Data Streams
    Johannes Gehrke
    Mirek Riedewald
    SIGMOD Conference(2003), pp. 40-51
    Automating Layout of Relational Databases
    Rakesh Agrawal
    Surajit Chaudhuri
    Vivek R. Narasayya
    ICDE(2003), pp. 607-618
    SWIM: Scalable Weakly-consistent Infection-style Process Group Membership Protocol
    Indranil Gupta
    Ashish Motivala
    DSN(2002), pp. 303-312
    Preview abstract The group membership protocol described in this paper (SWIM) has since been implemented by Uber and has been in use in their production infrastructure for several years as of 2016(!). It is a key scalable building block of Uber's Ringpop, and is referenced prominently on their engineering blog: https://eng.uber.com/intro-to-ringpop/ View details