Before you participate in the Kaggle competitions, it is great to be informed about its ranking system and how you can progress faster. This progression system is pretty straightforward as it is intended to track the growth of the users on this platform. On your journey to the top, you will learn a lot, earn medals, make new friends and form teams. As you progress you will compete in live challenges and engage in discussion boards for real-world problems. There are certain competitions and company projects restricted for “Kaggle Masters” only; you need to unlock that level to witness another world of data science and machine learning work and glory. Welcome to the arena, this is yours to take; you need to prove your worth.
Let’s start by looking at the categories that are based on your expertise and interests.
The progression system on Kaggle is specifically designed to cater for different levels of expertise. There are three main categories – Discussion, Kernels, and Competitions with their own rules of progression and rewards. Advancing from one performance tier to another is all up to you. You need to ascertain the level of expertise and the kind of data that you can handle with confidence as you move along.
Within each category, there are different performance tiers that a Kaggle contributor achieves depending on the quality and amount of work that they do. In each category, there are five performance levels with the lowest performance tiers being Novice and then Contributor. The mid-level performance tier is Expert while the highest levels are the Master and the Grandmaster. To get different performance tiers from Novice to Contributor a user needs to deliver work consistently, learn from their mistakes and gradually improve their rankings.
It is possible to rank differently in each category such as a Discussions Master, a Kernels Novice, and a Competitions Contributor. The highest tier that you achieve in any category is what will be reflected on your Kaggle profile just under your avatar/picture. So how are tiers awarded to a contributor? Pretty simple it depends on the number of medals that a contributor earns in that category.
This is the tier you automatically join when you sign-up for Kaggle. This is the starter category that every Kaggle user begins from and then advances from there.
This is the performance tier that a Kaggle user has completed his profile, have engaged with the community and made a kernel submission. To achieve contributor status, you must engage with the rest of the community through discussion forums and cast at least one UpVote and explore the platform to its entirety. Some of the information required to be a contributor is to add your occupation, personal bio, location and the organization you are affiliated with. Additionally, a contributor must also make at least one comment, cast one UpVote, run a single script and also make one submission to any competition. Lastly to be a Kaggle contributor one must verify their account through SMS. Kaggle only allows one account per user and regularly removes users with multiple accounts or profiles.
This is a performance tier that a user achieves once he has completed a substantial body of work in single or more categories of expertise. When a user finally attains the expert level in a category, he is also ranked in the general Kaggle ranking of the particular category. To become an expert in discussions a user needs fifty bronze medals, two bronze medals for competitions category, and five to become an expert in kernel category.
This is the second highest level on Kaggle and takes a lot of effort and commitment to attain. This level is a demonstration that you are taking full advantage of this platform to sharpen your data science skills. If you are at the master level in the competitions category, you get the express permission to take part in prestigious projects reserved for top-tier Kaggle contributors and only accessible to Kaggle masters. To achieve competition masters level a user needs to have two silver medals and at least one gold medal. To become a master in the kernels category, a user needs a total of ten silver models, and to become a master in the discussion category, user needs fifty silver medals or a total of 200 medals altogether.
This is the ultimate recognition of users’ efforts and the highest performance tier level in Kaggle categories. To get to this level, one needs to demonstrate a high level of commitment and consistency and be an active member of the community. To become a grandmaster in the competition’s category, you need at least one gold medal and five gold medals with one being solo.
To become a grandmaster in the kernels section, you need a minimum of fifteen gold medals, and to become a grandmaster in the discussion category, you need fifty gold medals and a total of at least five hundred medals.
What do you need to earn medals?
When it comes to the Kaggle, medals refer to the standardized way of rewarding and recognizing ones’ effort across all different expertise offered on the platform. One medal is awarded for one accomplishment, a widely popular kernel, a helpful or insightful comment or an impressive competition result.
Medals in the Competition Category
To earn a medal in the competition category, you need to get top competition results. How many medals a user is awarded depends on the size of a particular competition. Keep in mind that Getting Started competitions, Playground Competitions, and In-class Competitions do not earn any medals. You only earn medals in the competitions category through featured competitions.
If a competition has up to 250 contestants, you need to rank in the top 40% to get a bronze, top 20% to get a silver and top 10% for a gold medal. For competition with 250 to 500 contestants, you need to rank in the top hundred to get a bronze, top fifty for a silver model and top ten in addition to be in 0.2% for a gold medal.
In a competition with more than 1,000 teams, you need to rank in the top 10% to get a bronze medal, top 5% for a silver model and top 10 in the leader-board (in addition to 0.2%) for a gold medal.
Medals in the Kernels Category
Medal in this category is awarded only to the popular kernel depending on the UpVotes that a single kernel receives. Keep in mind that al UpVotes do not necessary go towards the medal. When calculating medals to award, votes on old posts, self-votes, and votes by novices are excluded. To get a bronze medal, a Kaggle user needs five votes while silver medals need twenty votes. To get a gold medal, you need a minimum of fifty votes.
Medals in the Discussion Category
You are awarded medals depending on the popularity of the topics you post and how many comments a post gets. This is measured using net votes which are determined by subtracting DownVotes from the UpVotes.
Kaggle Points and Kaggle Rankings
The points and rankings show where one stands as a user, if you feature on the live leader-board it is an indication that you are among the best in this data community. Your profile on Kaggle will show your current rank as well as the highest rank you have ever achieved on Kaggle. To be ranked on the live board, you must rank as expert or higher on that category in question.
Medals and tiers are permanent representations of your achievements as a data scientist on the Kaggle platform. On the other hand, points are temporary as they ‘decay’ over time. The objective of Kaggle points is to keep Kaggle rankings both competitive and contemporary. The formula for point decay is e-t/500 – where t refers to the number of days elapsed after a point is awarded.
In a competition, Kaggle points are awarded depending on the performance in a competition based on how many people for a team. Just like medals, there are no points awarded in InClass, Getting Started and Playground competitions.
When it comes to the Kernel category points are awarded depending on their popularity. Every UpVote is worth a point and decays depending on when the vote was cast. Lastly, discussion points are determined by subtracting the total number of down votes from the total sum of UpVotes. Both down votes and UpVotes decay depending on the date that they were cast.
How fast you will advance on Kaggle solely depends on your effort and commitment to do as much work as possible. Rising from a novice to a contributor level can take a few hours or a day, depends on how fast you complete the required criteria – complete your profile, make one competition submission, write a single comment, cast one UpVote and run a script.
Rising from the contributor level to all the way up to be a grand master is another story. It is a learning process that requires consistency, active participation in discussions and engaging in competitions regularly. Currently, there are only 94 Grandmasters in the world. This perfectly indicates how much effort it takes; there is no shortcut, it is all about hard work and learning from each competition you undertake.
Most of the masters and Grandmasters on Kaggle have been on this platform for more than two years. Stanislove Semenov; who currently ranked 2nd on Kaggle confesses that his four-year journey on this data science platform has been an incredible learning experience. He explains that his intention was not to rank highly but to learn, the ranking just came naturally the harder he worked and the more regularly he contributed to the platform.
I’ve recently published a book Kaggle for Beginners, I hope you will enjoy it.