October 22, 2020
We all consume media on a daily basis from a variety of sources. That is almost entirely unavoidable. This means that we place a level of trust in those that provide our news that the information supplied to us is truthful, reliable, and valid. Is that really the case though? (spoiler: the answer is no).
In light of the 2016 referendum on the UKs membership of the EU, I set out on a project to write a classifier that would be able to determine whether a news article in the British media has a pro-remain (in the EU) or pro-leave bias. I aimed to use the most modern data science tools at the time, which would surely give me the best results. Right? ...Right?
In the end, I built something that sucked, and I’m going to tell you about the journey of excitement, horror, pain, misery and finally acceptance that I went on while developing my models.