• Bubble
  • Bubble
  • Line
Data Science and Machine Learning new package for old ideas
Andrei Clinciu Article AUthor
Andrei Clinciu
  • 2018-08-14T20:57:00Z
  • 5 min to read

Data Science and Machine Learning are to extremely hot topics at the moment.
All of the bigger companies want to have Data Scienceitsts and Machine Learning engineers or experts work for them. Many courses offer information for these two.

While they may seem like two distinct world, Data Science and Machine Learning ARE actually extremely closely related to eachother.
If you've recently looked at the average salaries for Data Science Experts and Machine Learning Engineers you can easily conclude that this is a GREAT and NOBLE job.
Salaries that go beyond $110.000 per year on average. Comparing these to other software engineering salaries we can conclude that these two jobs are even more in demand due to the current hypes.

Most people claim that Data Science and Machine Learning are "new concepts". The demand has increased due to the fact that many businesses have failed to grasph the importance of Statistics in day to day life. This is true because Statistics aren't taught as they should be which leads to big problems. This is then the primary reason why Data Science and Machine Learning are quoted so high and why newbies with almost no experience can get pretty good salaries in comparison with other jobs!

While it's true that the job titles Data Sciencetist and Machine Learning engineer are new and most people (80 to 90%) have less than 10 years of experience, these two distinct jobs have existed for some while under different denominations.

We now mainly use computers to

  1. Gather the Datasets
  2. Analyze the data
  3. Make predictions on the data


Data Science and Machine Learning are concepts that have existed for a long while.
We now have the computational power and frameworks together with the tools necessary to make this tedious job seem easier to approach.

The History of Statistics and Data Science

Data Science is just a new name for Data Analysis or Statistical Analysis which has been around for ages, Statistical data has been collected and 'analyzed' for more than 2000 years.
Writing has been invented not for poetry or to describe but for business needs. There are written records even older like the Tartaria Tablets (5500 to 5300 BC). Ancient Summerians (3100 BC Mesopotamia) devised writing because they wanted to keep track of their sales and products. THis can only mean that prehistoric statistics had an important impact on the world.
The 18th century was a turning point for Statistics or "Data Science" as we know it. Most countries began implementing statistical systems to keep track of information. This is also when most staistical implementations took place.

Modern statistics has been perfected between 1900 and 1950. This can only mean that "data science" has existed for some time now. The problem is mainly that statistics has not been taught in schools at it should be. Statistics should be something all people must learn throughout their first 12 years in school.
Not only the formulas but how to gather information. Which information to gather and how to make sense of information.
Businesses today face big issues due to the simple fact that they;ve failed to grasph Statistics and their importance.
Whenever I mentioned statistics I didn't imply "graphics" we are shown in newspapers or on polls but the whole sciencetifical and mathematical aspect of it.

Machine learning

But surely Machine Learning is an extremely modern notion you might say. It surely is, since "Artificial Intelligence" can only be attributed to the time of Charles Babage's workings on his mechanical "computer" which was used guess what, again for statistical mathematics! https://iase-web.org/documents/papers/icots3/BOOK2/B1-14.pdf

The first mention of machine learning occured in 1959. Almost 60 years ago. Arthur Samuel said the following:
Machine Learning: Field of study that gives computers the ability to learn without being explicitly programmed. He also wrote a whole paper describing "Some Studies in Machine Learning Using the Game of
Checkers" a few years later he made a followup with recent progress.

Machine Learning and Artificial Intelligence have existed for over 60 years. Various implementations where done by many great programmers, they wheren't aware of this nor was there any unified framework.
Since 2010 we have seen more unification as the processes seem to converge to 1 unified study of Machine Learning.

What must Machine Learning and Data Science experts do?

Python and R are now trending with Data Science and Machine learning. They are trending because of the plethora of packages available to help a Data Sciencetist or Software Engineer cope with making sense of the data and implementing sane solutions.

I recommend anyone into Data Science or Machine learning to consider dwelving into Python and R.
Python has the most packages available for Machine Learning and it's truly worthwile.
Even if you don't plan on using python day in day out it will still pay off big dividents at the end of the year.
For example I'm using Elixir as my main programming language nowadays. I connect Elixir with Python for all my Data Science and Machine Learning activities, this way I don't have to search for similar packages in Elixir and I use the best out of both worlds.

Data Sciencetists and Machine Learning experts have to analyze and train systems with huge data sets.


Why are companies in such dire need of such people?

The reason companies are in a dire need for such people is that there is a big whole in the market.
Companies have started gathering data since data IS the currency of the world.
They gather data for everything that occurs inside the company, outside of it and everything their customers do.
This data needs to be analyzed and certain systems need to make choices based on it.
Because there is too much data people alone would never be able to cope with it. The solution is to hire experts that can find similarities and give solutions out of this data.

Making predictions on data must be done with care so that the predictions affect the bottom line on a positive side.

People are constantly saying that Data Science and Machine Learning is a fade and that it will pass.
This is true, it is a fade that has been going on for years and it will pass when the internet will dry out.
Untill then, the need for experts in such domains will soar.

The only problem at the moment is that these terms are pretty new for job titles and there aren't really many people with senior experience.
The majority of the people hired have between 1 and 3 years of experience.
On the downside you can never know if the person who is working has enough experience.
But on the positive side if it's someone who has a lot of software development and statistics knowledge then it doesn't matter if he's new to the field since it will work out perfectly for companies.


What if you're a small company and want to start using Data Science and Machine Learning?
There are basically 3 options, the last one being the best

      1. You either search for existing tools and try to do things yourself

  1. You externalize this to a 3d party company which will most likely cost a lot of money
  2. Or you just find a consultant who has the basic knowledge to collaborate and help you automate.


Want more information about how Data Science can help you? Feel free to contact me and we can setup a meeting where we can discuss the benefits and how you can tap into Data Science and Machine learning to improve your profits.

Ideas and comments

Andrei Clinciu
Andrei Clinciu

I'm a Full Stack Software Developer specializing in creating websites and applications which aid businesses to automate. Software which can help you simplify your life. Let's work together!

Building Great Software
The digital Revolution begins when you learn to automate with personalized software.

Find me on social media