Need help working on data science projects



Hi All,

I am new to Analytics Vidhya and Data Science and this is my very first post on this site. I am a working professional working on different databases from last 8 years and switched to data science just few months back. I did data science course with jigsaw academy.

I have around 7-8 big data/data science projects in my organization but I am not sure how to approach these problems.

One of my problems is below:

I have a dataset with millions of data points. There is a KPI that gets calculated based on different levels in the data. For eg: KPI Cost per Car gets calculated for Dealer, Model, Damage and multiple other drill down levels.

As per the current approach of guided analytics, a user has to drill down multiple levels to get the exact root cause for rise in the KPI value. We would like to change this approach by being a little creative and using data science to tell them the exact problem right at the click of a button.

I know how we can achieve this manually by traditional SQL methodology but I need some help in knowing which algorithm can work best in these types of problems. Finding exact root cause of a problem from millions of data points.

Any suggestions will be really appreciated.



Hi @winter022,
Better you can try “cluster analysis” for these kinds of problems.