Paul Owe

This large-scale project seeks to uncover characteristics and structural properties of the FSL Twitter network. A graph processing platform to create a dashboard of influential Twitter profiles and a geo-visualization tool are implemented as part of the big data solution

This notebook guides you through the process of building a k-means clustering model for market segmentation using BigQuery ML, iterating on the model to develop a version that creates the optimal number of clusters, and using cluster analysis evaluation methods such as Davies-Bouldin index

Walkthrough of how to use Pandas & Matplotlib to analyze and answer business questions about efficient advert timings, and further uncovering optimization opportunities using association rule mining to generate frequent patterns

As part of the NSERC CREATE Training Program, I attended the public seminar delivered by postdoctoral trainee, (and former undergraduate teaching assistant) Amin Omidvar where he spoke about how to design the last layer of a neural network for various supervised machine learning tasks.

As part of the 76 projects funded by the Academic Innovation Fund (AIF) at York University, I adapted UPenn's open source flask application to display location-based open educational resources on a map.

I virtually completed my internship with KPMG's data analytics consulting group. As part of the internship program, I worked with a business client to complete data quality assessments, conduct research in knowledge mining, prepare visualization dashboards, and presentations.

I conducted a social network analysis to understand the structural properties of networks on twitter. I used tweepy to test and integrate with twitter’s REST and Streaming APIs.