| Technical details | |
|---|---|
| Submit file | my_vivino.ipynb |
My Vivino is an online marketplace. We have a vast wine database, and we have 27 million users, mainly in North America.
One of our leading services is a wine recommendation system. It starts to be a little old. Rules-based. We are selling wine based on our customer visit/research.
You've just finished a meeting with your product manager. Data science is the future, and our competitor my_wine.com has invested a lot in it.
Following your suggestion, my_vivino's CEO has decided to allocate a budget to move forward on the data science project you've proposed.
Which project?
What are the success criteria?
- During our next meeting, you will have to show us some data (plot? report?) of what you've been building.
- Impact on the business. We need to make our customers happy.
What to expect?
- Slides presentation.
- And the DevOps team should be able to push it to production.
You've heard the CEO will be joining the meeting. It's a reminder if you do well, you can quickly expect the promotion you are expecting.
What is to be a Data Scientist?
- Data Collecting / Cleaning
- Data Exploration
- Data Visualization
- Machine Learning
- Communication
You will have to prove yourself in each of these. We are confident you've already done it! :)
Where to find the data? Any where. Scrap Vivino / Bevmo / Delectable / Wine-Searcher / ...
It's an open project. You are free to find something you find will be useful to do some data analysis on:
- rating system?
- comments system? detect most valuable comments?
- wine suggestion?
- wine classification?
- wine quality?
- size of the market analysis?
- suggestion base on a meal recipe?
Reminder, it will be one of your portfolio projects. You can find a lot of different ideas. Plagiarism is not tolerated in the company either here. :-) If you don't find enough information on wine, why not moving to the Whisky market, Beer market, or Coca? any beverage market? :-)
You are in data science, so your blog post needs to report as a "scientific experiment"; you need to write your assumption, every step of the experiment, and the conclusion. Usually, with an opening to what you follow, what to experiment next?
The scope is vast. You will find your idea :)
How will this project be graded?
- Your code
- Your code quality
- Your slides quality
- Business impact
- Your blog post (You can inspire yourself from this fascinating blog post on Kubernetes.)
How To Use Jupyter In Docode