My Data Science Lab

Predicting Bike Rentals¶

While I was was taking the Data Scientist stream of courses on DataQuest.io I worked on several excercise projects for gaining hands on experience on Data Wrangling, Exploratory Data Analysis and some basic Machine Learning algorithms.

In this post I have captured the work that I did in one such excercise project: Predicting Bike Rentals

In the previous post I described the problem and the dataset, and performed a comprehensive EDA. In this post we will develop Machine Learning models that will learn from the seven weeks of sales data and predict product demands of the following two weeks.

We will begin by reading-in the preprocessed data from the previous post.

Predicting future demand based on historical sales data is one of those things that businesses have been doing for time immemorial. It is a critical part of the Inventory and Supply Chain Management.

A typical solution would look at the sales data, for example the number of items sold, for a certain period in the past and use statistics to project the demand in the near future. Reasonably good solutions would include seasonal variations and the more sophisticated ones might even take into account the demographics data and the market trend.

User Segmentation is an important requirement for Targeted Digital Marketing and Customized User Experience. While a lot of new ways of Customer Segmentation are being used lately, thanks to the explosion of online social media, the traditional way of segmenting users based on Demographics is still very relevant. Knowing just the gender and age/age-group of a person can tell a lot about their interests, inclinations and their propensity of buying certin kinds of products.

This post is the second and final part of TalkingData Mobile User Demographics two-post series. The first part can be found here: TalkingData Mobile User Demographics - Part 1.

In this post I will describe the following three models and compare their performances based on the evaluation metric that was used in the competition for scoring.