Welcome to Tariq’s Homepage!

Currently, I am a Ph.D. Student of Department of Computer Science at the University of Rochester. Previously, I worked as a faculty member in the Department of Computer Science and Engineering at the Bangladesh University of Engineering and Technology (BUET).

Earlier I completed my M.Sc. and B.Sc. in Computer Science and Engineering (CSE) from BUET.

Research Experience

Currently, I am working in the ROC-HCI lab led by Prof. Ehsan Hoque, Associate Professor, CSE, UR. I have joined the PARK project, where I am trying to detect Parkinson’s from videos containing facial expressions.

During my M.Sc. Thesis and most of my academic career, I have worked on big data analytics, cloud computing, and distributed systems. In my M.Sc. Thesis, I worked on deriving a scalable and federated technique of PCA, which is applicable to any arbitrarily large dimensional datasets. I would like to highlight that, an article titled Fast, scalable, and geo-distributed PCA for big data analytics, based on my M.Sc. Thesis was published in the Elsevier Journal Information Systems in the 98th volume of May 2021.

I also have experience working on large-scale social media data analytics and a manuscript titled UACD: A Local Approach for Identifying the Most Influential Spreaders in Twitter in a Distributed Environment was published in the Social Network Analysis and Mining (SNAM) journal. In this work, I have derived a local algorithm of identifying the most influential spreaders on the Twitter social network using our own designed user-attributed code decomposition (UACD) method, which provably can produce more accurate results ($\mathbf{12.5\%}$ on an average) in $\mathbf{175\times}$ reduced running time. Our derived method has been empirically manifested to be effective in large-scale social graph analysis on a distributed platform.

In addition, I also have explored some other fields such as Deep Learning, Network security, and Bioinformatics. For example, in the case of network security, I have designed a model with hierarchical attention for the host intrusion detection system (HIDS). In bioinformatics, I proposed a novel multi-layer convolutional network for protein function prediction. The manuscripts for both projects are close to completion.

Research Interest

My primary research interest is Heath Data Analytics. I am also interested in Big Data and Social Media, Distributed Systems, Cloud Computing, and Federated Learning.