Project: A Data-driven Approach for Improving the User Experience of Internet Users
Our lives are heavily reliant on Internet-connected devices and services. However, to deliver the desired user experience over the Internet, network operators need to detect and diagnose various network events (e.g., disruption, outage, misconfiguration, etc.) as well as resolve them in real-time. We have developed an Internet-wide measurement infrastructure that collects performance metrics (e.g., latency, jitter, throughput, packet loss rate, signal strength, etc.) from vantage points deployed by real users (mobile phones, WiFi access points, etc.) at regular intervals.
We currently collect data from millions of source-destination pairs, and we have plans to expand our footprint to billions. We are looking for data science interns to process the collected data (sparse) to train robust learning models that can accurately infer various network events in real time.
One selected candidate will receive a stipend via the DSI Scholars program. Amount is subject to available funding.
Faculty Advisor
- Professor Ethan Katz-Bassett
- Department/School: EE/SEAS
- Location: Morningside Campus
- The goal of the Networking and Systems Research Group is to improve the reliability and performance of Internet services. To understand the problem space, we look at the needs of operators and providers and conduct detailed network measurements. Based on what we learn, we design deployable systems to improve the Internet and services that run over it. In the last few years, we have worked with Google to deploy our techniques to speed up connections for all users, with Facebook to control the paths used to direct traffic to its two billion users, and with Microsoft to build the system that monitors user connections to steer them around performance and availability problems.
Project timeline
- Earliest starting date: 05/27/2019
- End date: 08/02/2019
- Number of hours per week of research expected during Summer 2019: ~40
Candidate requirements
- Skill sets: Python (Pandas, SciPy, Scikit-learn, etc.), SQL
- Student eligibility (as of Spring 2019):
freshman,sophomore, junior, senior, master’s - International students on F1 or J1 visa: eligible
- Additional comments: Student should have the drive to make a real-world impact.