Data Mining with an Eye to Algorithmic Bias

This workshop will guide you through simple first steps for doing data analysis, specifically text mining with Weka. Weka is an open source machine-learning tool. We will be replicating the work of Mike Thelwall in his paper on Gender bias in machine learning for sentiment analysis. By the end of the workshop you will have applied a sentiment analysis technique to a data set with two different techniques and be able to compare the bias of each technique. From here, you are in a position to both continue using text mining to analyze data, but also to have seen a concrete example of algorithmic bias in machine learning.

This is an introductory workshop, with no coding or other pre-requisites. Please bring a laptop or sign one out from the Loans and Returns desk.

Presenter: Francisco Berrizbeitia is a developer for the Concordia Library. He currently works on applying text mining and machine learning techniques to document classification for a variety of purposes.