The work of the Data Scientist, as we know, requires a lot of skills, including statistics and math concepts, visualizations skills, and programming knowledge. For a beginner, all of these requirements could be a deterrent. In fact, one might think of leaving without even taking the first steps.
In this article I describe my experience and how I began to take my first steps in the immense jungle of Data Science. Basically, by following nine simple steps, and with a lot of patience, you can become a (good) Data Scientist.
Before Starting the Journey of the Data Scientist
Before we get into the world of Data Science and then put the tips I will describe into practice, there is a bigger battle to be fought.
The first real battle is against anxiety. Yes, because, taken by too much work, you might think you have to do everything immediately in the best way.
This was, in fact, my first impact with Data Science: I wanted to know how to do everything, even before starting.
In carrying out the work of the data scientist, I personally felt anxiety in these cases:
- a dataset too complex to analyze — while applying many preprocessing techniques and the like, the dataset still remained large and complex;
- a Machine Learning algorithm that takes too long to train — waiting for the result is nerve-wracking; an impending deadline and still partial results;
- a piece of code that doesn't work - it's frustrating to spend hours and hours trying to get code to work or find a solution to a problem, sadly without success;
- too many concepts to learn (especially regarding statistics and maths) and little time to study.
If you have experienced anxiety while starting the Data Scientist journey in other situations, feel free to comment on this article. I will be happy to update the previous list with other anxiety situations as well.
And then I thought, this situation couldn’t go on. I stopped to reflect and in the end I developed nine tips to put into practice to start the journey of the data scientist.
The first way to begin the learning process is to keep calm. Theoretically it seems obvious, but practically, perhaps it is the most difficult thing to apply.
Keeping calm is a virtue of the strong. It involves strong self-control, endless patience and great temperance.
The key to staying calm, for example before learning a complex statistic concept, is to withdraw from the situation. Just think about the present and imagine having your whole life available to learn.
Another important aspect to become a good Data Scientist is to have a good knowledge of the topics covered. The only way to be sure how to proceed with data cleansing, analysis, etc., is to study.
There is no other choice: I have to study.
I cannot leave with my car without starting the engine first. And the same goes for the work of the data scientist. I can't train a model if I don't have some theoretical knowledge first.
The study techniques can vary: you can take a manual and study, you can read articles about it, etc. Personally, I prefer to read interesting articles and test examples, because only by practicing I can learn something.
From my experience, the only way to progress in knowledge of a subject is to set aside time for study, whether it is an hour a day or an hour a week. The important thing is that this time exists. Surely the time spent studying is not time wasted, on the contrary, it helps to make progress quickly.
When working on the computer, we often forget that we are not alone. Many times I spent a lot of time searching for the solution on Google, sometimes with success, sometimes without success.
In reality, someone else may already have the solution to a problem ready, which seems insurmountable to us.
Over time I have learned that there are numerous communities of data scientists, very willing to help me in times of difficulty.
For example, there are Data Science communities on Quora, Stackoverflow, Reddit, Linkedin and in general on all social networks. Just ask!
In order to combat anxiety, another fundamental element is the organization of work.
Over time I have learned that a piece of paper is worth more than a thousand documents on the cloud, precisely because it allows me to touch (and focus with my eyes) all the things to do.
Basically, I took an old diary and started writing down all the things to do month by month. I have considered each Data Science task a single thing to do and to learn. I have highlighted each completed task in yellow. What a satisfaction to see the diary gradually turning yellow!
Thanks to the agenda, the my knowledge of the various topics has gradually increased, because everything is immediately under control and it is easier to organize the work to be done today and tomorrow.
Often overwhelmed by the enthusiasm of doing so many things, I started too many Data Science activities or projects and eventually fell into the anxiety of not succeeding.
From these experiences I learned to measure my strength, balancing the workload to what I can actually do and the time I have available.
This is not about having or not having the skills to start a new Data Science project or a new business. Instead, it is a question of measuring one's strength, to avoid rushing things and falling back into failure.
Closely related to the previous point is not expecting too much of yourself. To tell the truth, I have sometimes found myself demanding more of myself than I could give at that moment.
There are moments in life when you perform more and others when you progress more slowly. The important thing is to never stop. All the same, even if for a while, you walk more slowly.
For example, I can't expect to make beautiful
Mike Bostock visualizations if my skills don't go beyond Excel!
Continue reading on Towards Data Science