Biography

I started conducting research in machine learning as a PhD student in the School of Computer Science at the University of Sheffield in November 2021, advised by Professor Haiping Lu. Before joining UoS, I had worked in the industry as a software engineer for five years, where I was in charge of developing software, bug fixing, and data visualization.

I completed the MSc degree as the first honor student in computer engineering, artificial intelligence, from the University of Kurdistan in 2013. My Master’s thesis focused on dimensionality reduction of big data through feature selection to improve classification accuracy. To achieve this, I came up with a framework to apply ant colony optimization algorithm to the filter approach without using any learning algorithms. I published two highly cited papers in high-ranked journals on my research work. Performing the Master’s thesis opened up new perspectives in my career to apply artificial intelligence in solving real-world problems. Therefore, after graduating from university, I wrote a research proposal for Tehran's Molecular Biology Research Center where challenged me with cancer detection/classification on high-dimensionality microarray data with mislabels or unreliable samples. My next paper was published on the proposed method in the Neurocomputing journal.

I never lose my passion for developing a comprehensive feature selection tool. Therefore, I joined the University of Kurdistan to devise an open-source software package named Universal Feature Selection Tool (UniFeat) by implementing 30 well-known and state-of-the-art feature selection methods by considering filter, wrapper, and embedded approaches. Fulfilling this academic project has been a unique experience for me and the the tool has been published in the journal Neurocomputing.

For previous methods lose their performance over time, my experiences in different projects from university to research center to industry led me to pursue a PhD program in machine learning at the University of Sheffield. I am currently investigating the efficient feature selection methods for integrative multiomics analysis to tackle the curse of the dimensionality phenomenon of these data types.