Take home data challenge - samples

data_science
guide
reporting

#1

Hi all,

I am a fresh graduate and actively applying for data science positions. Since I have no professional experience yet, I would like to know if there is any public repo, blog or book where I can find guidance on how to prepare for take home data challenges. Apart from building predictive models (common challenges), I am interested in knowing:

  1. What are other types of common challenges?
  2. What are the main key points to include in a report. For instance, should I include all steps followed in data cleaning, all tried and failed prediction methods, …?
  3. What are the hints to prepare for these kind of challenges

I will also love to hear from anybody who has gone through this


#2

Hi @kthouz

First for type of challenge well Vidhya is one, Kaggle is a well known one more diverse. Challenge go from optimisation, to supply chain, fraud etc… There are the week end hactaklon as well in many places, they are good and some companies are recruiting there, for example predictive maintenance, health related prediction could be there. For me the last week end it was about microorganism in blood stream !!! all a program great experience and good prize :slight_smile:

Point 2, well if you speak of competition the metrics are the king !! so reports forget only during hackthaklon you have to so a presentation about your solution do not bother with algorithms etc… only end results

Point 3. This is personal, build a team if you want to achieve in small time frame. Second know the data transformation, feature engineering and some models. Most of all practice before !!! a lot some time read read read !!!

Perhaps other people have other advises.
Alain