Lessons from building a spam filter

4-Sep 2018 - 19:15

Are you interested in using machine learning to leverage knowledge from your data? Do you want to know more about the possibilities and especially the pitfalls you need to be prepared for, before you start your own project using machine learning? Then this is the presentation for you. Johan, who have studied the field of machine learning for many years, will showcasing the process of building a spam filter to catch negative posts in his web forum CityPolarna.se. The algorithm used is an LSTM RNN network, but as you will be aware of after this talk, it is the data that makes most of the difference.

About the speaker:

Johan Broddfelt work as a software developer at FrontIoT ApS in Copenhagen. He is also one of the organizers of the network Barrel AI and co-founder of the community CityPolarna.se. He has a Batchelor degree in Computer Science and has been developing administrative web applications for the last 20 years.