There are many of you who are working as data analysts or are doing some job where you have to deal with a lot of data. Being in the field of data, you must have heard of data science, machine learning or artificial intelligence. Fascinating job prospect, interesting problems and skyrocketing salaries must have grabbed your attention, and you must be wanting to become a data scientist by now. Believe me, if you deal with data in your job in any way, you are just a step away from getting into the data scientist job. Here, I am assuming that you know a little bit about data science and have learnt some algorithms, now you are striving to work on real time problems, build your resume and move to a proper data scientist job.
If you work on data for analysis, cleaning, visualization or any other task, all you need to do is :
- Think of a problem statement around the data
- Apply some machine learning algorithm
- Get some prediction results which you think can help your organization even at a smaller level
- Create a nice presentation and show it to your manager in company, which will have at least:
- Problem Statement
- How solving that problem using data science will help your company/department
- Your approach to the problem and application of data science algorithm(s)
- Results and recommendation
Your manager may approve or reject the idea, that’s a different issue. He/she may not even look at the presentation and may kill the idea then and there itself. But in the entire process, you have completed your mission. You have achieved what you wanted.
By following the above process, you actually completed the life cycle of data science project. You started by defining a problem statement and ended up with some results which you think may help your company. Now, what were the outcomes of this process:
- You worked on real time data
- You sailed though a data science project
- You learnt and applied data science algorithms
- You have got a data science project to write about in your resume
- You can apply to data science positions by showing couple of such projects that you carried out
- You can talk about the project in details and with ease because you have good domain knowledge about the data and project as you have already been working on it as data analyst.
Now, let me give you some examples about what you can do with the data:
- If you are working in a bank and work on transaction data, can you think of predicting about which translations are fraud/money laundering transactions which are not and help your bank avoid heavy fines?
- If you have fraud transaction data, can you think of clustering and find in how many types of fraud transactions are there and find their characteristics and help your company to prevent such transactions by applying some strategies?
- If you are working in HR/recruitment department, can you think of creating a model which can predict about attrition of employees? Which types of employees stay long with company and who quit early? Can you distinguish their characteristics and help company hire more loyal employees?
- If you are working in HR/recruitment department, can you create a time series of job opportunities in your organization and predict how many jobs of what skill-set will be required next year and helping company align their recruitment strategy accordingly?
- If you are in telecom sector and work on customer data, can you try to find clusters of profitable customers and help your company to devise strategies to attract more of such customer clusters by offering good plans.
- If you are in supply chain department and work on logistics data. Can you think of optimizing it in such a way that company’s cost is reduced considerably?
Supplement these real data projects with participating in various Hackathons organized by different analytics platforms. Gain as much practical knowledge as possible by applying multiple appropriate algorithms on the data given in such hackathons.
Believe me, IT WORKS. I can tell you this with confidence because I have opted for this formula to get into data science field. Good thing is that Data Science is in very nascent stage now. Organizations are realizing the value of data science and need good data scientists. Demand for data scientist is far exceeding the supply. Organizations value knowledge and experience. By showcasing these projects and expertise that you gather while working on them, you can impress your interviewer. It does not matter if your idea and data science project could see the light of the day, but you got the knowledge and experience of a data science project which you can talk about.
I would just end this by saying that hard work pays in the end. Just follow the advice given in the blog. Initially, you may not be able to think of any problem statement associated with data you work on. But, more you learn about data science algorithms, work on different practice problems, read articles about how data science is solving problems in your domain, participate in hackathons, more you will be able to think of a practical problem statement about your data and how solving it using data science can help your company. At the end of the day, it would be a win-win situation for both you and your organization. The worst-case scenario of it is that you will end up with a data science knowledge and a project on your resume.
Now when you are done with getting and doing the projects, your next task is to create a resume with all these projects. You can read the detailed article here about how to write a resume which stands out and can fetch an interview call. Next task is to go and crack the interview. Read the article on how to crack a data science interview with ease.