The Ultimate Battle Ground of Data Scientists Data Science Competitions to Compete, Win, Practice, Learn And Build your Data Science Portfolio!

Analytics Vidhya was, is and continues to be the most important source of my education.The hackathons are exciting, the discussions forum is becoming more and more active every day and the content on the website is excellent. No wonder it has grown so much in the last few years.

AV is one of the best knowledge portal for Data Science I know about and certainly the best in India. It caters to the need of all sorts of people, from beginners to well experienced professionals. The articles are in-depth in nature and cover wide-variety as well which is very difficult to find else where. Discuss portals help people to clarify their doubts while hackathons give a flavor of real world experience. Thank you AV and keep up the good work.!

Analytics Vidhya hackathons are an excellent opportunity for anyone who is keen on improving and testing their data science skills. The portal offers a wide variety of state of the art problems like – image classification, customer churn, prediction, optimization, click prediction, NLP and many more. I have participated in many hackathons and they have been a significant contributor in improving my machine learning skills. These hackathons have real time submission and evaluation feature and you can always feel the competition heat on the leader-board. Hackathons also provide an immense opportunity for learning from other’s, as winner’s approach and code is shared at the end of the hackathon. I won the Barcelona water demand prediction hackathon and it boosted my confidence and network significantly.

I learnt a lot by participating in various Hackathons conducted by Analytics Vidhya during last three years. One obvious benefit is given a dataset I can intuitively understand what works best which I apply in real life problems efficiently. My strategy in any Hackathon is to understand the dataset, feature engineering and cross validation. I try ensembling, hyper parameter tuning or stacking only if I have time. I believe feature engineering and cross validation is more important than your choice of algorithms.

I started my journey in analytics in August 2014 through an online course. Then came AV which gave a newbie like me a platform when my first article was published.Thereafter the association with AV has grown to AV being my guide and mentor. There is no dearth of analytics resources on the web but when you get it from experts in the Indian analytics space with a clear business problem in mind-there is nothing like it.The easy to understand language used to explain the most complex of subjects in data science is one of the cornerstones of the AV blog.The one place I know I will find answers to my questions and gain knowledge is AV. In the language of Data Science AV is an ensemble of many experts which bring out what is best for the community. :)

Right from college, I liked to participate in competitions and Analytics Vidhya helped me compete amongst the finest people. Now, even as I'm working as a full time data scientist, I still like to give it my best during these short yet intense competitions as it helps me think out of the box and get the best out of myself. I usually start with looking at the data and checking the distribution of features and target variables which involves a lot of group by commands. Once I settle for an initial feature set, I train a basic model to see where I am in the leaderboard. From there on it can go to completely revamping the model to use some new methodologies or tuning the model and features.