Corporate Experience by Chetan Singhal, Data Analyst in an MNC. A detailed description of the learnings and workflow.
Hello everyone!!!
This is Chetan Singhal, a postgraduate from IIT Bombay in Applied Statistics and Informatics and a graduate from Hindu College in Statistics. I am working as a Data Analyst in an MNC.
I know that the field of Data Analytics is emerging a lot. So is the number of students trying to explore and start a career in this domain. There is no limit on the knowledge in this domain that one can gain. But it’s important to understand the requirements. So, I am sharing my professional experience with you. I hope to give you insights about what are the tools and techniques that are being used practically.
Firstly, coming to the tools, Python and Excel are the ones that I am using for doing the analysis and PowerPoint for making presentations. Try to be familiar with Excel tools like Pivot Tables, Vlookup, Macros, visualizations as these will be required in every domain of analytics.
Coming to Python programming language, it has become the most popular one to analyze the data, build models, and for doing more advanced data science stuff.
In Python, I have worked on the following things:
- Pandas: A library in Python to slice and dice the data and helps in the summarization of the given dataset to a great extent. It uses DataFrame objects to store data and do operations on it efficiently
- Lists, Dictionary, and tuples: One should be proficient in using these data structures in Python.
- Scikit Learn: This is a package that has all the machine learning algorithms and other related stuff
- XgBoost: This is one the most widely used algorithms for classification and regression and it’s a package also in Python through which we can implement it.
I have talked about the tools, but what exactly the workflow is, I will explain now.
- Data comes to us in raw format from clients. Thus, a lot of pre-processing is required to make it clean. So, the first step is to understand the data, field names, and the pre-processing.
- Then, we summarise the data and try to see the distributions of different variables in the dataset. We also try to identify different trends and patterns, if any.
- Then comes the feature engineering part, where we create the features for our models based on the descriptive analysis done earlier.
- Finally, we train different models like Random Forest, XgBoost, neural nets, and many others and validate them on test data.
- Finally, when the models are built, there are different ways to cater to the needs of the clients like creating worklists, and then the models are deployed.
My experience has been great learning till now. I would encourage the ones reading this to first know your areas of interest and keep learning new things. There is still a long way ahead. Wish you all a great and successful journey ahead!!
Like this Data Analyst Corporate Experience, you too can share your experiences with us. For any suggestion, please reach out to us on LinkedIn.
Find some of the resources that helped us here.
You can create an impact by talking about your interview experience. Please fill this form .
You can read other articles here.
Cheers and Best!