Michael Rodino

he/him #datascience. #rstats. #pydata BLM. data scientist. Ana nerd. ex-ppc’er. current ppc x seo friend. hockey stat fan. works @Comcast. opinions are my own.

[Michael Rodino @michaelrodino]

Started a new job last week. Amongst all of the new imposter syndrome weight - let me tell you - nothing beats the feeling of the first successful query from a dataset. Maybe I DO know how to write SQL and work with data!

[Ron Berman @marketsensei]

Companies invest a lot in analytics - but are these investments valuable?

@IsraeliAyelet and I studied ~1,500 online retailers and found that using a descriptive dashboard increased their weekly revenues by 4%-10%.

>>
#MarTech #BigData #Analytics #ecommerce #DataScience

[Michael Rodino @michaelrodino]

You can never be too careful. Do it for the kids.

[Chelsea Parlett @ChelseaParlett]

An oldie but a goodie 🎃👻

[Michael Rodino @michaelrodino]

“True” Outliers = an opportunity to ask the product owner & engineering team what they were thinking when they let that data in here.

[Chelsea Parlett @ChelseaParlett]

REPEAT AFTER ME: We get rid of outliers when their extremeness indicates they are not part of our population, we do not get rid of them solely because they are extreme.

[Michael Rodino @michaelrodino]

It’s… so beautiful…

[Chelsea Parlett @ChelseaParlett]

It’s the time of the semester again!

Here is my mildly embryonic concept map of the algorithms I cover in my Intro Data Science course 🎨👩🏻‍💻

[Michael Rodino @michaelrodino]

@imightbemary Either that or “I’m going to talk through this to make sure we are on the same page, but please let me know if I should skip ahead or if I am missing something.” Opens the floor to make sure we agree on fundamentals of a problem and catch my own mistakes!

[Michael Rodino @michaelrodino]

@imightbemary I regularly try to stop myself before I go into an explanation, and ask anyone to interrupt me if I’m explaining something they already know. It’s not perfect but it has helped a lot.

[vicki @vboykis]

The more programming languages I learn, the fewer strong opinions I have about any one language other than just as a tool to get something done.

[Michael Rodino @michaelrodino]

Great thread!

[Andrew Dessler @AndrewDessler]

One of the things you learn as a scientist is the ability to look at a plot and think, "that just doesn't look right." That's the feeling I got when I saw this plot that Lomborg is currently pushing.

[Michael Rodino @michaelrodino]

@PhDemetri Are you going after one type of goat or will you diversify?

[Michael Rodino @michaelrodino]

One of my favorite things is singing parody versions of songs around the house at full volume. This is the best one for this song I’ve seen yet. I don’t care how small of a group will get it.

[vicki @vboykis]

This is ten percent luck
Twenty percent skill
Fifteen percent concentrated power of will
Five percent pleasure
Fifty percent test/train/cross-validation split,

[Michael Rodino @michaelrodino]

Help me #DataScience #statistics Twitter, you’re my only hope…?

[Michael Rodino @michaelrodino]

It only took me one week off in Maine and extra time on #stats Twitter to want more good stats textbooks to brush up on fundamentals again. Recommendations please?

[Michael Rodino @michaelrodino]

Casella Berger is in the list, but I’ll take any applied regression books - and I’ll also take recommendations for good independent shops to buy from and support.

[Michael Rodino @michaelrodino]

@data_mesh_learn This was actually, very helpful. Thank you!

[Michael Rodino @michaelrodino]

If I say data mesh will someone explain it to me?

[Michael Rodino @michaelrodino]

@ChelseaParlett We love a good explainer.

[Michael Rodino @michaelrodino]

It comes for us all.

[vicki @vboykis]

Made a quick data career progression flowchart for anyone confused by all the titles in the field.

[Andreas Madsen @andreas_madsen]

Our new survey on post-hoc interpretability methods for NLP is out! This covers 19 specific interpretability methods, cites more than 100 publications, and took 1 year to write. I'm very happy this is now public, do consider sharing.
Read https://t.co/03TmDZRsfy. A thread 🧵 1/6

[Demetri (is over at the other place too) @PhDemetri]

There is a lot of benefit to be gained from being at least half statistically minded in applied machine learning settings.