CSE graduate student earns NSF Graduate Research Fellowship for research on data mining
Tara Safavi’s research focuses on scalable and adaptive data mining algorithms using tools like hashing and sampling.
CSE graduate student Tara Safavi has been awarded an NSF Graduate Research Fellowship for her research in data mining on graphs, time series, and sequences.
Safavi graduated in April 2017 from U-M with Highest Distinction and High Honors in computer science, and is now finishing the first year of her PhD, advised by Prof. Danai Koutra. Currently, her research focuses on scalable and adaptive data mining algorithms using tools like hashing and sampling.
Safavi’s research was recently nominated for a best paper award at the IEEE International Conference on Data Mining. Her project “Scalable Hashing-Based Network Discovery” was motivated by the growing need for scalable data analysis, and seeks to address the problem of efficient network discovery on many time series. Evaluations using real data showed that her method constructs networks nearly 15 times faster than baseline methods, while achieving comparable network structure and accuracy.
Outside of research, Safavi enjoys teaching. She served as an IA for three courses as an undergraduate (EECS 183, 280, and 490) and volunteered for two coding organizations aimed toward teaching youth (UM WISE Girls Who Code and Seven Mile Coding in Detroit).
Safavi has also earned the Google Women Techmakers Scholarship, and was the recipient of the College of Engineering Marian Sarah Parker Prize and an Outstanding Research Award from the EECS Department as an undergraduate.