Soukaina Filali Boubrahimi

Soukaina Filali Boubrahimi

Assistant Professor of Computer Science

Utah State University

Biography

Prior to joining USU as an assistant professor in the Department of Computer Science of the College of Science, I obtained my Ph.D. and M.S. in Computer Science from Georgia State University in 2020 (under the supervision of Dr. Rafal Angryk). Throughout my PhD and data science internships in Expedia and LexisNexis, I was particularly interested in interdisciplinary research that involved designing new methods for mining high dimensional time series data for the task of prediction. I recently received the title of “Outstanding Teaching Awardee” of the year 2019 and “Outstanding Research Awardee” of the year 2018 from the computer science department of GSU. I was also featured as the next exemplary Second Century Initiative (2CI) Fellow in the university: full interview.

Interests
  • Time Series Mining
  • Spatiotemporal Data Mining
  • Rule Discovery
  • Space Weather Research
Education
  • PhD in Computer Science, 2020

    Georgia State University

  • MS in Computer Science, 2019

    Georgia State University

  • MS in Software Engineering, 2015

    Al Akhawyan University in Ifrane, Ifrane, Morocco

  • BSc in Computer Science, 2014

    Al Akhawyan University in Ifrane, Ifrane, Morocco

Teaching

 
 
 
 
 
Fall 2021 - CS 5080/6080
Time Series Data Mining
Aug 2021 – Dec 2021 Interactive-broadcast (Huntsman Hall 270)
 
 
 
 
 
Spring 2021 - CS 6675/7675
Advanced Data Mining
Jan 2021 – May 2021 Web-broadcast
 
 
 
 
 
Fall 2020 - CS 5080/6080
Time Series Data Mining
Aug 2020 – Dec 2020 Web-broadcast

Courses Description

Advanced Data Mining (3 Credits)
This course provides a deep dive into advanced topics in mining texts, graphs, time-series data, vector datasets, and frequent itemset and association rules. The lectures will provide students with a sufficient foundation to apply data mining techniques on massive real-life data repositories using Python. Students will gain hands-on experience in the chosen aspect of the data mining area through the completion of a major data mining project. Topics covered include Node2Vec/Word2Vec models for text and graph embedding, vector space models, time series classifiers, representation learning, data reduction, and association rule mining.
Time Series Data Mining (3 Credits)
This course provides a broad introduction to state-of-the-art research on data mining, machine learning models, and statistical pattern recognition on time series data. The goal is to learn how to apply, inspect, and evaluate different mining techniques on time series data using Python. Topics covered include time-series representation learning, Fourier and Wavelet transform dimensionality reductions, similarity search, classification, visualization, and frequent patterns mining. Additional coursework is required for students enrolled in the graduate-level course.

Projects

Mining Minimal Shapelets Set

Mining Minimal Shapelets Set

This project addresses the high numerosity problem of mined shapelets issue by mining the minimal set of discriminative shapelets for time series data

Rule Mining

Rule Mining

This project explores the idea of time series rule mining using Allen’s Interval relationships

Time Series Generation based on Auto-encoders (TiGA)

Time Series Generation based on Auto-encoders (TiGA)

This project augments time series datasets using time warped autoencoders

Publications

Quickly discover relevant content by filtering publications.
(2021). Spatiotemporal event sequence discovery without thresholds. Geoinformatica, (25), 1, pp. 149–177.

(2020). Evaluation of Hierarchical Structures for Time Series Data. NA, pp. 94–99.

(2020). Machine learning in heliophysics and space weather forecasting: A white paper of findings and recommendations. arXiv preprint arXiv:2006.12224.

(2020). Multivariate time series dataset for space weather data analytics. Scientific data, (7), 1, pp. 1–13.

(2019). A scalable segmented dynamic time warping for time series classification. NA, pp. 407–419.

Contact