Applied Machine Learning*
Are you intrigued by Machine Learning but don’t know how to actually use it? This talk will focus on a specific case, solving a large scale Entity Resolution (De-Duplication) problem with an open source Support Vector Machine (SVM).
Machine learning is a powerful (and very cool) approach to solving certain sorts of complex problems. Unfortunately, the available documentation tends to be abstract and academic. This talk will be a fairly detailed tutorial on applying machine learning algorithms to real-world challenges. We’ll cover the following:
- What’s the problem we’re trying to solve?
- What is a SVM and why is it an appropriate solution?
- How do we actually implement a SVM to solve our problem?
– Creating the feature vector
– Translating raw data to ML-friendly data
– Training a model
– Applying the model
- How did this work out for us?
I have virtually zero speaking experience. I've given a few tech talks at work, but that's it.
Clayton McClure is a Sr. Software Engineer at Rentrak, working on back-end data processing for medium to big scale data in the entertainment industry.