UROP Project

*** Anomaly Detection Research and Development for Real-World Datasets

anomaly detection, python, programming, data science, artificial intelligence, machine learning

Research Mentor: Amanda Lovett,
Department, College, Affiliation: Center for Ocean-Atmospheric Prediction Studies, Arts and Sciences
Contact Email: alovett@coaps.fsu.edu
Research Assistant Supervisor (if different from mentor):
Research Assistant Supervisor Email:
Faculty Collaborators: Shawn R. Smith
Faculty Collaborators Email: smith@coaps.fsu.edu

Looking for Research Assistants: Yes
Number of Research Assistants: 1
Relevant Majors: Computer science, data science, scientific computing, computer engineering, or other related disciplines.
Project Location: 2000 Levy Avenue, Building A, Suite 292
Research Assistant Transportation Required: FSU bus system provides transport between main campus and Innovation Park
Remote or In-person: Partially Remote
Approximate Weekly Hours: 5-10, Flexible schedule (Combination of business and outside of business. TBD between student and research mentor.)
Roundtable Times and Zoom Link:
Not participating in the roundtable

Project Description

If you are looking for a research-centric project that will teach you how to work with real-world data, identifying different kinds of anomalies, and implementing machine learning models, this is the project for you. The Marine Data Center (MDC) at FSU is looking for a student who is willing and eager to engage in a research project that requires investigating existing data and developing code, specifically implementing machine learning frameworks, for working with this data.

The MDC receives weather and ocean data from numerous research vessels on a daily basis, which results in an extensive dataset spanning back 20 years. There are several circumstances in which data is considered anomalous, such as instrument malfunction or an extreme weather event. These anomalies are reflected in the data itself, which data analysts often have to manually identify and record. Considering the sheer quantity of data the MDC works with, this can be an extremely time consuming affair.

The goal of the project is to lay down the foundation for future anomaly detection work at the MDC. The student should have an interest in machine learning, programming, and experimenting via trial and error. The student is expected to actively try different pre-existing machine learning frameworks to determine what systems may be effective for our datasets.

Research Tasks: The student will be responsible for researching, investigating, and developing machine learning systems for use on real-world datasets. The student will work with both data analysts and software developers to obtain a better understanding of both data and programming concepts. The student will need to study current anomaly detection frameworks and apply them to in-house data to determine which systems are effective, which includes setting up training/testing sets. The student will be expected to report on their progress to superiors on a weekly basis and communicate should any issues arise.

Skills that research assistant(s) may need: The ability to learn new programming languages and/or frameworks, basic understanding of data analysis, paying attention to detail, and communicating effectively with team members will be essential. Any level of experience with past programming projects, specifically those involving Python will be beneficial, but not necessary. Experience with Linux or using a command line interface is desired, but not required.