“The unprecedented explosion in the amount of information we are generating and collecting, thanks to the arrival of the internet and the always-online society, powers all the incredible advances we see today in the field of artificial intelligence (AI) and Big Data.” (Forbes) Organizations around the world are leveraging this explosion of data to solve some of their biggest business problems. Banks can better predict loan defaults, retailers can improve customer experience, and much more. 

With so many questions to answer, what are some of the most common machine learning problem types that come up while building out AI systems? Jake Shaver, Special Projects Manager at DataRobot, walks us through four problem types in this installment of AI Simplified. 

1. Classification
2. Regression
3. Time Series
4. Anomaly Detection 

“It’s important to understand which problem you’re solving as each problem can use different models, have different accuracy metrics, and other problem-specific parameters that you need to account for.” — Jake Shaver

Watch the video below to learn more about each problem type along with common use cases:

{% video_player “embed_player” overrideable=False, type=’scriptV4′, hide_playlist=True, viral_sharing=False, embed_button=False, width=’1920′, height=’1080′, player_id=’11281140402′, style=” %}

Ready to learn more about machine learning problem types? Check out these items below:

• Reducing Readmissions with DataRobot at Steward Heath Care

• Accelerating Time Series Analysis with Automated Machine Learning

• Advances in Fraud Detection with Automated Machine Learning

About the author

Ashley Smith

Meet Ashley Smith