{"id":6488,"date":"2019-09-16T22:11:43","date_gmt":"2019-09-17T05:11:43","guid":{"rendered":"https:\/\/careerkarma.com\/blog\/?p=6488"},"modified":"2020-05-23T00:21:25","modified_gmt":"2020-05-23T07:21:25","slug":"how-to-use-kaggle-to-learn-data-science","status":"publish","type":"post","link":"https:\/\/careerkarma.com\/blog\/how-to-use-kaggle-to-learn-data-science\/","title":{"rendered":"How to Use Kaggle to Learn Data Science"},"content":{"rendered":"\n

It\u2019s 2019, and data science is one of the most attractive career options<\/a> out there. Data scientists get to spend their days working with bleeding-edge tools building models and performing analyses for everything from predicting rainfall to trading the stock market. The typical data science team is staffed with a diverse array of people, from talented newcomers to grizzled stats PhDs, lending a kind of hybrid vigor to the whole field. 
<\/p>\n\n\n\n

And there are more books, tutorials, courses, and bootcamps for data science<\/a> than you can shake a stick at. But the fact remains that getting into data science can be an intimidating prospect. For this reason, the more possible entry points you have, the better. You can never know what explanation or demonstration will finally bring home a concept you\u2019ve been struggling to understand. 
<\/p>\n\n\n\n

One resource for learning and practicing data science which I\u2019ve used myself, and which I\u2019ve heard of other people using, is Kaggle. But since I\u2019ve never seen anyone write up an explanation of how to do this, I decided to create my own. 
<\/p>\n\n\n\n

What\u2019s a \u2018Kaggle\u2019?<\/strong><\/h2>\n\n\n\n
\"\"
Kaggle is a well-known machine learning and data science platform. <\/figcaption><\/figure>\n\n\n\n

Kaggle is essentially a massive data science platform. It gathers in one place a huge number of public datasets, most of which have been sanitized and made ready for use in analysis. By itself this is pretty significant, as data gathering and cleaning is a huge part of the data science workflow. 
<\/p>\n\n\n\n

Kaggle also hosts a number of ongoing competitions, where individuals and teams work to solve data science or machine learning problems under a set of constraints. Winning or just placing highly in one of these contests has become a big enough deal that people routinely put it on their resumes and LinkedIn profiles.
<\/p>\n\n\n\n

Because Kaggle users publish notebooks that are freely available for anyone to browse, adapt, and use, it has become an extraordinarily rich source of code for data science and machine learning projects. 
<\/p>\n\n\n\n

And while Kaggle is most certainly not a proper social network, it has grown in popularity to the point that it has many active messaging boards and a substantial community. It\u2019s a great place to pose questions and to hash out ideas. 
<\/p>\n\n\n\n

Learning By Imitating Good Kaggle Projects <\/strong><\/h2>\n\n\n\n
\"\"
Copying the moves of someone more skilled than you is a time-tested means of getting better. 
<\/figcaption><\/figure>\n\n\n\n

One of the things I did immediately after graduating from the Galvanize Data Science Immersive was to spend some time each day exploring and carefully studying Kaggle notebooks which utilized data science skills I wanted to master.  
<\/p>\n\n\n\n

There were a couple of things I was looking for. I chose notebooks that:
<\/p>\n\n\n\n