Alif Munim

I'm a fourth-year undergraduate co-op student studying Computer Science and Psychology at Ryerson University. My aim is to achieve a deeper understanding of human perception and reasoning using machine learning. I am primarily interested in its applications in mitigating media bias and aiding biomedical professionals.

I'm incredibly fortunate to be supervised by Dr. Dafna Sussman as a computer vision research assistant at Ryerson's Maternal-Fetal Imaging Lab, located at the iBEST institute. Prior to this, I was supervised by Dr. Eric Yu at the University of Toronto as a research assistant in natural language understanding and business automation.

Academic Experience

Classical vs Deep Learning Methods for Image Segmentation of Small Brain MRI Datasets

A review of approaches to overcome the challenges that exist in biomedical image processing; low segmentation accuracy, small datasets, and low resolution. I worked with Dr. Dafna Sussman, Daniel Nussey, and Rachita Singh at Ryerson's Maternal-Fetal Imaging Laboratory to compare u-nets, fully convolutional networks, and gradient-boosted ensemble models for the segmentation of brain metastases in <100 3D MRI scans.

Email Speech Act Classification for Task Automation

A collaborative effort between the University of Toronto and IBM's Center for Advanced Studies (CAS) to extend prior work in speech-act theory and bring it into the realm of deep learning. More specifically, we utilize earlier work by Carvalho and Cohen (2005) defining illocutionary points in "email speech acts" to build taxonomic word embeddings for semantic relatedness. I had the pleasure of being supervised by Dr. Eric Yu during my time contributing to this project. Our work culminated in the launch of IBM Watson Orchestrate, a productivity assistant that won a CES 2022 Innovation Award.

Industry Experience

Building Intelligent Enterprises with IBM Watson AIOps

IBM Cloud Pak for Watson AIOps is an intelligent IT solution for businesses. Using AI for predictive log analytics and anomaly detection, Watson AIOps automates complex IT processes and vastly reduces risk. As a software engineer intern on the Watson AIOps team, I took on a lead role in building efficient data pipelines for data cleaning, feature extraction, and AI training.