Introduction:
This is a short exercise to
test your ability to construct and evaluate machine learning models. You should
produce a machine learning model and an evaluation metric. The goal of this
exercise is not to produce a state-of-the-art machine learning model. Which
model you use, and how you evaluate, is up to you. The choice of model is not
important (although we will assume that when you choose a model, you understand
what it is and how it works). Your solution should be simple, but sensible: you
should be able to explain why it tests something of impact to the problem.
Dataset description:
In the
"training_sales.csv" you will find the hourly sales of a store in the
US. The following is the description of the dataset:
Date: Date
Time of the Sales
Value: Sales in
cents
In the
"training_traffic.csv" you will find the hourly traffic of the same
store in the US. The following is the description of the dataset:
Date: Date
Time of the Traffic
Value: Traffic data
in person measured using store sensors
We
are looking to identify the sales and traffic per hour for the following month.
You can use any external data you wish (PS: you might want to look for holidays
and national days).
The
dataset has some missing points, this is deliberate to understand your
assumptions about the missing data, and how you handle them.
Submission: Please submit the
following:
1-
A
report discussing your methods, and the final results
2-
All
your code
3-
Any
external dataset used
Please
feel free to use either R, Python, or Scala, and please feel free to use any
out of the box functionality. We are mainly looking for your thinking process,
and how you would approach the problem.
Get Free Quote!
267 Experts Online