Creating Your Data Science Project — Data Scientist Perspective
What are the important aspects of your project
As a Data Enthusiast, creating your Data Science Project is something that could be either easy or hard. Although, I know for sure; Nobody at first who would evaluate your Data Science project, so you did not know is your project good enough in the eye of the company.
Here I would share a few tips of what the company would want to to see from your project — If your interviewer is a Data Scientist:
Problem Solving
What question did you ask? Is the problem significant enough to solve with Data Science project? How do you approach the problem? What are the steps you take to tackle the problem? And so on.
These are the main questions what a Data Scientist would ask from your Data Science project. Whatever it is, your project needs to solve the problem or at least shown that you have to try to solve it.
VentureBeat shows that 87% of Data Science project would never make into production; the reason is various, but most are no real solution introduced from the project.
If the company current asset unable to solve most of the problem, why they want to hire you if you cannot show the potential as a problem solver?
Business Centric
You might show the creativity, but is your project actually viable and useable in the real-world business situation?
Maybe the Data Science project sounds cool, e.g. Predicting the number of rain-fall that fall into your pants — It is a challenging project, and if actually did it I would be amazed. But, is it useful in the business situation?
If from the beginning, the project showed that there is no place at all for this Data Science project to apply to any kind of business, then it is useless. Academia, maybe, but not the company.
I didn’t say that your Data Science Project should be related to the business you are trying to apply to (This is the best, however). What I try to say is your Data Science Project has a business value in any business? That is the main point.
Clear Concept
The clear concept here means that you know what problem you want to solve, the features you used, what approach you take, and the precise steps to achieve it. Just like a blueprint, you know inside out of your project.
Never put a half-baked project in your portfolio, especially when you want to have an interview. We Data Scientist know what project you did with a clear concept from the beginning to the end and which project could just be taken from the Kaggle or someone GitHub.
If you are lucky to get invited for an interview and the interviewer asks you to present your project, that would be a disaster if you do not have a clear concept.
Have Fun
It is your project, shows that you are enthusiastic and happy to create the Data Science project.
Write the explanation in your notebook cells like you have fun with a precise lining of how your creativity produce this Data Science Project.
For me, Having Fun in your project is the most important aspect you could show to the company.
Lastly,
My tips: If you have the project in place already, ask the Data Scientist you know (or me) for reviewing your Project. They might suggest some amazing name for your project as well.
I hope it helps!