Data Science - Part III - EDA & Model Selection

For downloadable versions of these lectures, please go to the following link:

This lecture introduces the concept of EDA, understanding, and working with data for machine learning and predictive analysis. The lecture is designed for anyone who wants to understand how to work with data and does not get into the mathematics. We will discuss how to utilize summary statistics, diagnostic plots, data transformations, variable selection techniques including principal component analysis, and finally get into the concept of model selection.

[hunhwasong]

Hello Derek. I am a data analyst in a software company in S.Korea. I literally don't know how to thank you uploading these invaluable videos. I am teaching myself how to do data analysis step by step.

[MadhuKumarKilli]

Hello Derek,
First of all i would like to convey a very big THANKS for these wonderful videos. Your videos are so descriptive and so easy to understand that i am referring them for my pre-interview preparations. I haven't come across of any other Youtube videos which are so structured and provides a comprehensive explanation. I am so grateful for you for sharing these videos on Youtube. These are very helpful for the budding Data Scientists / Data Analysts like me.

[upalchowdhury]

Best Data Science Lectures that one can use in real life i mean in job.

[dbs]

Great stuff! These set of videos could have been the best had it not been for the audio! Thanks for sharing!

[senthilramalingam]

Thanks for sharing it. It creates lot of insights on how to analyse data in step by step process with concrete methodology.

[mariav]

Thank you for posting these lectures. They are great!

[suhardisjahrial]

Thank you...this is suitable for me who want to get an introduction of data science - without a strong background in statistics

[chadrob]

One little feedback. @29:30 where you explain about Location and Variability, you have explained Location as 'central value of your variable.' But Mode is not necessarily the central value, especially when variable has a lot of skew. I am sure you must know all this but it might have missed in the flow. Needless to say, this series of videos is greatly structured and very valuable even for person who knows a things or two about data analysis.

[vinothvvkumar]

Thank you, it's an amazing video, very well structured and explained

[MilitanT]

Man I wish the audio quality was nice that I can watch at 2X. Increasing the volume makes the noise unbearable :|

[lxk]

Thanks for uploading this!! it's really helpful. I'm wondering if it was possible that you could share some code which you used for visualizing. Those plots look really nice!!

[jeffwantf]

Hi Derek, thank you very much for these videos

[researchdesk]

Thanks Derek, your series is really good

[tomash]

Good video but I did not like being too short with PCA. There was no intuition behind loading coefficients explained. Also for transformation it would be nice to see step by step example but as an overview good enough

[aerospaceconstructor]

it would be great to get subtitles in english ))

[cassianrozario]

really good series especially R and R Studio
may I have your personal email to discuss a project with you
thanks in advance

[sauravprakashgupta]

Too much noise in the video.
Though I want to watch. I can't continue