I am a Computer Scientist, an Interdisciplinary Researcher, and a Data Scientist.

I very much enjoy research; especially research that serves the public good.

I am not a big fan of the term "data science", but I am told this is what I do!

I appreciate working on the data that matter for the society and I passionately study them.

I believe in the power of AI; Good or Evil is in us not in AI.

We only need to pick the right battle and fight.

. . .

Sponsored by NSF and NASA grants, I have been working as a Graduate Research Assistant at the DMLab in Georgia State University. In an interdisciplinary team of Computer Scientists and Solar Physicists, my research objective is to help provide insight into space weather activities through research-to-operation and operation-to-research efforts. Using machine learning and statistical tools, I focus on (1) data-driven pattern extraction, (2) prediction of occurrence of the solar events with sever impact on our infrastructure, and (3) follow-through on translating back the findings into the host domain, Solar Physics.

Due to the high-dimensionality and heterogeneity of data coming from an array of ground-based and space-borne observatories, a typical project in this area requires engineering of complete pipelines of data integration, preprocessing, curation, training, and evaluation. These are compute-intensive tasks that I carry out thanks to the in-house computing resources of the DMLab.

. . .

AREAS OF INTEREST:

Machine Learning, Deep Learning, Image Processing, Time Series Analysis, Performance Verification, High Dimensional Data, Imbalanced Data, Spatiotemporal Data

RESEARCH TOPICS:

• Model Evaluation:

  • Evaluation of Salient-Object Detection with Fine Structures

  • Evaluation of Categorical and Probabilistic Classification

  • Robust Sampling of Rare Events

• Spatiotemporal Analysis:

  • Prediction of Space-Weather Events

  • Detection and Segmentation of Scientific Events

  • Task-Specific Feature Extraction and Optimization

• Multivariate Time Series Data:

  • Statistical Feature Extraction

  • Time Series Clustering