Learning Machine Learning13 Dec 2016
For the last six years, I’ve been working professionally in the field of data science. My work has ranged from helping clients understand their website usage at a digital agency, building a monitoring tool for Tumblr bloggers, leading user acquisition at a venture-backed startup to finally work in data analysis and engineering at Intel Security. Through those different positions, there is one constant: I am intensively extracting value from raw data in order to drive decisions.
During those years, I’ve been doing lots of data engineering, analysis and statistics. However, I rarely had to use machine learning algorithms in order to answer questions or resolve problems. It’s crazy how far you can go with basic statistics, quality data and the right questions. People often underestimate this.
As I progress in the field of data, I’m starting to notice many opportunities where my work could be enhanced with machine learning. Like many jobs, a part of the data analysis work is prone to be automated with artificial intelligence in the coming years. Machine learning has already automated some of the data science tasks.
In the next few months, I want to focus on learning machine learning. I’ll be deep diving in the topic. I will write about my learnings in the process and I will share as much as possible with you. In the past, I’ve tackled with machine learning here and there. I took a few basic classes, skimmed through books and worked with colleagues to implement some machine learning models. I have a decent understanding of the different algorithms. I am not starting from zero but I have a long way to go.
Why am I learning in public?
The main reason is that I want to share my learnings with you. There are many studies that demonstrate the benefits of teaching in the learning process. I’ve been really inspired by people like Nathan Barry who teach everything they learn as they go.
To start with, I will take a hands-on approach. because I know this works best for me. I am not overlooking the theory behind machine learning. I’ll get to that at some point in my learning process. The main resource I’ll be starting with is a book called Python Machine Learning by Sebastian Raschka. Raschka shares a lot of knowledge on his blog and I like his style of writing. He has many public Python Notebooks where he drops tons of knowledge about data science. The Python Machine Learning book was recommended to me by a few people I look up to. It will be my starting point.
Are you learning machine learning too? Do you have some recommendations? Please drop me a note.