Introduction
The key to learning anything is practice, practice, and practice. The higher you go on the learning curve the more complex your practice exercises get. Do you remember the leap from your first coding exercise to building side projects over the weekend? It all builds on incremental practice.
One of my biggest challenges as a self-taught analyst(for the most part) was building my data science projects portfolio so I began with building my little data science projects.
Let us talk about how you too can plan and build your little first project which could be your gateway to a new job or an addition to your portfolio.
Identify a case/problem
Your choice of a case to work on can potentially be solving a problem, bringing a new perspective to existing phenomena or proving an unknown phenomenon. Take your time to research on scenarios you are interested in, what you can potentially do with them and whatever you choose, aim to do a comprehensive data analysis on it.
Tip: Always pick a problem or a scenario that correlates to something that you are passionate about building that way you will stay motivated to work through the entire project.
Finding a data set
Now that you have a case, you need to set out to find data that relates to the case. There are so many free data sets across the internet, however, if you want to take it a notch higher you might try to collect your data so that way you can learn what goes into designing a form that collects the right detail for your case. Otherwise, you can pick a sample data set from all the open data forums on the internet.
Tip: Kaggle has a great collection of datasets and lets you see what other people have done with these datasets, a great start for your little project!
Asking questions & Telling stories
Data Science centres on asking questions but not just any questions, you need to ask the right questions. In this step, you are going to formulate several questions that you will answer using the data set at hand and present these in a story flow to make sense of the data. One of the other common facets of data science is storytelling and using the answers to your questions you can write a great and compelling story to justify the case.
Tip: Write a script to prepare for this step, it will guide you on writing a great story and make your analysis straight to the point.
Examining Trends and interesting facts
While working with data, it is important to think outside the box and explore parameters from different categories to seek any unknown correlations and this step is also crucial in addressing bias and stereotypes in the data. At this step list down on all variants of test cases that can be applied on the data at hand and test each one while tweaking ideas.
Tip: This is a discovery process, so be open-minded.
Presentation
Now it is time to visualize and communicate your findings. First understand the audience you intend to communicate your findings to, as much as we love graphs, I recommend layering graphs with other graphics to captivate your audience.
Tip: Explore visualizations out of the traditional graphs and consider layering different graph styles to create captivating graphs.
Conclusion
Once you create your little data science project, share it with the world and add it to your portfolio. This can help someone in the industry and contributes to the wider knowledge that the data science community is building around the world.
Top comments (4)
hi! really nice article. Thx for sharing! what kind of project did you build?
Hello, thanks.
Check out a simple survey project I created and documented here youtu.be/2AyygaVGEpM
Do you have an example of a project along these lines? I dig the explanations but a quick example idea to get the gears turning could help.
Hello Marissa, check out a short survey project I created and explained youtu.be/2AyygaVGEpM