5 steps to promote your Kaggle account
Hey!
In this article, I want to share my experience on how to promote my account on Kaggle. In order not to be unfounded, here is my account. I would not say that I am Mr. Popularity on Kaggle, but still I have some experience. I want to share it.
What is Kaggle?
It is unlikely that you would have come to this page if you did not know what it is, but I will explain anyway. Kaggle is a huge sandbox for those who are learning or training to work with data based on real datasets that other users publish.
Why do I need a Kaggle account?
If you want to work in the field of data, you are an aspiring Data Scientist or Data Analyst, then a Kaggle profile is your calling card. Portfolio, if you will. It is the portfolio on Kaggle that will allow you to demonstrate your skills to employers, the ability to analyze and reason, and also demonstrate your code.
Why promote your account?
Everything is very simple. The higher your leaderboard account is, the more attractive it is for a potential employer and for other users with whom it is useful to share experience and skills.
Now that we have dealt with the basic questions, we can proceed to the specific steps that I compiled based on personal experience.
Unique datasets
If you carefully analyze the datasets published on Kaggle, then approximately 90% of them are data borrowed from somewhere. Data downloaded and provided by open sources. You can do this if the data is in open form, but few people will be attracted to such datasets, because everyone can get them oneself.
Much more valuable datasets are those that you have collected yourself. Of course, you can also do it manually, but here I rather mean the scrapper you wrote. Remember that firstly, we fill out a profile on Kaggle for a potential employer. Think for yourself which profile will attract him - the one where the dataset is downloaded from statista.com or the one to which the python scrapper file is attached.
For example, here is one of my datasets, where I parsed the Pizza Hut official website or this one, where I parsed all transactions for a certain date and attached a GitHub link with my code. I'm not saying that I did everything perfectly, I'm just giving an example.
By uniqueness, I mean exactly that. The very topic of datasets (medicine, cars, social data, etc.) is all really unimportant. Kaggle users are interested in absolutely any data, and they work with all data. In this regard, I believe that Kaggle has an ideal community.
When publishing a dataset, I recommend filling in all the necessary and required information to the extent that Usability becomes 10.0.
Kaggle Dataset Example
This is important not only in terms of the visibility of your dataset, but it is also important to treat the users of your dataset with respect and provide all the information for their convenient work.
Nicely designed code
As I said before, it is important to treat others with respect. Kaggle is a very friendly community and should not be changed.
Here is one example of data analysis done by @sadeghjalalian with my dataset.
Kaggle Notebook Ideal Design
Headings, beautiful design, code explanation in the comments. This is one of the best notebooks I have ever seen.
And here is an example of a bad notebook . I thought it would be wrong to put someone else's notebook in here, so I dug through my notebooks and found this horror. The complete opposite of the previous example. How could anyone even read this?
Kaggle Notebook Bad Design
Please don't do this!
Be sociable
This is important not only in terms of promoting your account, but also for you as a professional. Feel free to thank for the code for your dataset, for advice on how to optimize the code, for the recommendation on what is the best graph to build or how to do data cleaning more correctly. Also, don't be shy about making comments like this. It is also important for others to know someone else's opinion.
For example, I sometimes ask to see my code and see if it can be optimized somehow. They always help me with this, for which I am grateful.
Kaggle Discussion
Participate in discussions and help others. It is important not only to receive, but also to give help. Also thank the dataset authors at the end or beginning of your notebook. Just plain politeness. It won't help your promotion.
Use only new datasets
On Kaggle, in the Datasets section, you can find Trending Datasets. If you want to promote your account, get the title of Grand Master and a huge number of medals, then you should write code only for these datasets.
Kaggle Trending Datasets
If you are just training or studying, feel free to forget about this point and have fun.
Other users most often pay attention to Trending Datasets, which means that the faster you create a beautiful and useful dataset, the more people will see it. Everything is simple.
Participate in competitions
To be honest, 99.9% of the competition on Kaggle is too tough for me. For me, they are either too complicated, or I can not find a team for myself.
Kaggle Competitions Page
In any case, this is a great way to be in front of everyone and light up. Also, regardless of the results, participation will be an excellent line in the portfolio. Here I can only recommend one thing - do not participate alone. Working in a team will allow you to distribute responsibilities and work more productively.
The easiest and most reasonable way to look for a team is on Kaggle itself or on Reddit, where you can write a post that you are looking for a team for yourself.
Kaggle Community on Reddit
The community is not very active, but I think it is possible to find a team there.
Conclusion
These are the simple steps I picked up to promote your account on Kaggle. I have compiled a guide based on my experience. It is possible that you disagree with something or want to add something. Any criticism is welcome in our community on Reddit. I will wait for you there.
Good luck!