Data Science and a Tree

By Steven Crofts, Vice President at Kenfront LLC.

So, what does this picture of a couple of young Boy Scouts have to do with a technical and business topic like Data Science (and its cousins AI, ML and DL)?

It’s nearly 100% certain this tree went down due to strong winds.  But most trees are made to withstand strong winds. So, what  happened?  Look closely at the roots. They did not go deep. They are very shallow. Certain weather, soil and planting conditions can cause roots to stay only surface deep.  When the right elements of mother nature come together, it doesn’t take much to knock a big tree down.  This Scout Troop quickly picked up on the problem as they easily noticed the very shallow root system.

As I have talked to peers and customers over the last few years, it has become increasingly obvious that some employees, and the projects they’re working on, don’t have roots that run deep.  Is that why we see so many Data Science initiatives that are now being considered failures? 

I’ve been thinking about this lately, especially as the Data Science concept as it first came out, has expanded into projects like artificial intelligence (AI), machine learning (ML), deep learning (DL), and so on. This has led young college age learners and experienced analysts and data users alike to explore how they can become Data Scientists.  A common pattern usually emerges:

  • Find a website that lists all the skills you need
  • Attend one of the latest and greatest conferences on Data Science and other stuff
  • Hang out at the local meetup groups
  • Learn Python and R and whatever open source flavor is popular
  • Take some AI courses at your local college or maybe even online
  • Review statistics texts from your old college course(s)
  • Install Spark and Hadoop and whatever else sounds good
  • Find interest, free, large datasets to download and see what you can do with them
  • Find a patient mentor, who will likely remain a friend
  • And so on…

This type of list is pretty ingrained in our brains because that’s how we’ve been taught to learn and network over the years.  The end result may or may not be successful.  Even after you’ve gone through the laundry list of learning and networking, you still have to break through and get in the door somewhere and get that first official job of Data Scientist.

All this is getting me to the point of this article.

Several years ago I sat in a presentation that had an impact on me regarding this “how do I get my foot in the Data Science door” topic.  I know I’m forgetting some of the detail, but I remember the general concepts.

The point of the presentation I attended was that to get into this particular tech sector, forget the traditional learning model and try to build your resume by building a portfolio of Data Science projects.  Going at it in that manner will not only be impressive on a resume, but you will learn the right tools that help businesses get results.

Here are some steps to build that portfolio.

Pick a topic that you’re interested in

I remembered this one well, because it’s a theme I preach to my Boy Scouts. Find project and activities that you really like and have passion for — and then exploit and enjoy them.  Even for you data professionals, it doesn’t have to be a business topic. Just something that will hold your attention.  If you have a hobby, try that topic.  Something that occupies your time, like family, church or sports. Then see if there are publicly available datasets or if it is possible for you to easily build your own dataset. Poke around. Get creative. Ask your kids for ideas—they’re smarter than you think.

Think about your goals and then pretend you just posted the accomplishment of your major success on LinkedIn and maybe Facebook

This concept caught my attention because you don’t really post it now, maybe never, but it helps you visualize in your mind how cool it would be to say something in a LinkedIn post like, “Today I finished a very interesting project looking at census information over the last 100 years and compared death statistics based on 7 different regions of the United States to USDA data regarding primary diets for those 7 regions. The results are fascinating, with great visualizations. The USDA has asked my permission to post the results on their website and social media channels.”

As the second paragraph of the post, you summarize your findings and include all the technology tools you used. Wouldn’t that look a lot better on a resume than just a list of classes you took. And don’t just do one project. Find as many topics as you can where your passion comes through. 

Of course, at this point, you haven’t even done the project.  So far you are just visualizing success and what that might mean for your career.

Now, the tough part, you have to do the work

Don’t make the project so hard that you’re working on it a year later.  But don’t make it too easy either that potential readers say, “my 9th grader did something like that last week in a MIS class.”   Make sure the datasets are sophisticated enough that they are “real world” and that the tools you use are open source and often used in companies to do the analysis that’s needed. 

The purpose of this article is not to tell you how to do the work or how to find all the exciting results. You will have to figure out that roadmap on your own. Be inquisitive. Go outside your normal way of thinking and analyzing. Learn the technology you have to learn. Ask others for input and review.  Is there other data you could bring into your original plan that might make it all more interesting? Multi-source data is getting more and more important to get real insights.

As you go through all the learning and thinking and questioning, ask yourself: do I love doing this?  After a few weeks if you’re miserable, maybe you’ve discovered that this type of work isn’t for you.  But for now, let’s assume you’re going to love it!

Make that inner author in you show up and clearly write up the results (or create a deck if you prefer)

If you’re not a writer, you need to learn and practice. Part of this type of career includes being able to write a report on the results, present it in written or oral form to senior management and peers, and show you have the skills you need to keep moving up the ladder.

Use simple writing. Don’t oversell yourself. Let the results do most of the talking. Use charts, visualizations, stories to outline and present the results.  If the project lends itself to it, show a demonstration or use your datasets to give examples of results and then how those results appear on a dashboard.

Will all of these steps land you a big time Data Science, AI or ML position?  Maybe not.  But you’ll still be glad for all you learned and how it makes you a better employee no matter the end result.  

And your roots will be a whole lot deeper.

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Youtube
Consent to display content from Youtube
Vimeo
Consent to display content from Vimeo
Google Maps
Consent to display content from Google
Spotify
Consent to display content from Spotify
Sound Cloud
Consent to display content from Sound