Nan Zhou

5230 Fiore Terrace K102 · San Diego, CA 92122 · (858) 214-4671 · n2zhou@ucsd.edu

I am a master student at UC San Diego majored in Computer Science. I'm interested in big data and web development technologies. My past experience includes data mining, feature engineering, and full-stack development.


Experience

Software Development Intern

AI Labs, Didi Chuxing
  • Predicted the safety of drivers' driving behavior by using the driver's dangerous driving records and map information where driving behavior occurred
  • Used Spark and Hadoop to process TB-level data. Utilized statistical and word bag algorithms to construct structured features
  • Built machine learning models with XGBoost and LightGBM. The AUCs of prediction reached 0.803, and the Top 10% Recall reached 46.11%. Compared to the previous model, the AUC gained 3pp higher, the Top 10% Recall gained 5pp higher
  • Applied a patent as the first author. The structured features have been used in other projects
November 2017 - September 2018

Software Development Intern

  • Implemented and optimized rule updating algorithms for packet classification
  • Combined hash-based Tuple Space Search and decision-tree-based HyperSplit, realized the SmartUpdate packet classification algorithm, which firstly predicted the updating time based on overlaps among rules and then leveraged grouping and dynamic decision-making mechanism
  • Compared to traditional algorithms, the incremental updating speed of SmartUpdate is one thousand times faster and its throughput is five times faster
September 2017 - November 2017

Research Assistant

Network Security Lab, Tsinghua University
  • Utilized the emotional states of information propagators to predict the structural patterns of online diffusion
  • Implemented the hybrid sentiment analysis approach which combines supervised machine learning algorithms with the lexical knowledge to build emotion-labeled social network dataset, then performed two-step logistic regression classifications
  • Significantly improved the prediction performance compared to previous methods, from 0.694 to 0.749 in AUCs
  • Submitted two papers as the second author and were accepted by INFOCOM and ICACCE
February 2017 - June 2017

Education

University of California San Diego

Master of Science, Department of Computer Science and Engeineering

Computer Science Track

September 2018 - Present

Carnegie Mellon University

WitechLab, Department of Electrical and Computer Engineering

GPA: 95/100

June 2017 - September 2017

Tsinghua University

Bachelor of Engineering, School of Information Science and Technology

GPA: 92/100

August 2014 - July 2018

Skills

Programming Languages
Big Data Tools
Web Development Tools

Awards & Honors

  • Outstanding graduates of Beijing (top 5% of all graduates in Beijing)
  • Outstanding graduates of Tsinghua University (top 5% of all graduates in Tsinghua University)
  • Chang Dong Scholarship (1 out of 560, the highest honor in the Department of Automation)
  • Qualcomm Scholarship (top 1% in the School of Information Science and Technology)
  • 1nd Prize - Outstanding Student Research Training (SRT) project in Tsinghua University
  • 2nd Prize - Tang Hui Electronics Scholarship of China Instrument and Control Society
  • 2nd Prize - Student Academic Science and Technology Competition
  • 3nd Prize - Student Lab Construction Contribution Award
  • Scholarship for Excellent Technology Innovation - 2017
  • Scholarship for Excellent Academic Performance - 2016
  • Scholarship for Excellent Academic Performance - 2015
  • Scholarship for Excellent Volunteer - 2015