UPDATED: Video Added from the Conference!
Introducing Wale Akinfaderin > @WaleAkinfaderin < presenting “The Creative Art of Engineering Bespoke Features”.
Wale Akinfaderin is a Data Scientist at Lowe’s Company, Inc with expertise in machine learning, deep learning, statistical experimentation and general information theory. He has broad experience in implementing and extending ML techniques to solve practical and business problems.
In the past, he has built a data veracity model using machine learning and natural language processing techniques for international development with IBM Research and he was a 2018 Insight Data Science Fellow. He is an advocate of open source and democratizing learning and his writings on data science and machine learning have been featured on Hacker News, Data Science Central, Dataconomy, R Bloggers and IBM DSX.
He is a machine learning proposal reviewer for Cambridge University Press, the founder of Data Science Tallahassee meetup group, and a data science expert mentor for Insight2Impact which is funded by the Bill and Melinda Gates Foundation. He also advises startups in the data science and cognitive computing space.
Feature engineering is rudimentary to applying machine learning models to raw and messy data. From multiple imputations for missing data to filtering in feature selection for issues like information redundancy, crafting bespoke features can be used to understand the underlying patterns in a dataset, facilitate the machine learning processing pipeline and improve model evaluation metrics.
In this talk, we will explore feature extraction and representation techniques for numeric, categorical and spatio-temporal features. We’ll also highlight other advance feature engineering methods like non-linear encoding for linear algorithms, topological filters, data leakage and best practices.
Come check out Wale Akinfaderin’s talk at ML4ALL happening April 28th-30th in amazing Portland, Oregon! Get your tickets to attend here. For the schedule, our excellent sponsors docs for the conference, check out the ML4ALL Conference Site!