This is notebook is the fifth and last project of Udacity Data Analyst Nanodegree.
In this last project we're going to work with Exploratory Data Analysis and Explanatory Data Analysis, besides a few other topics about data analysis, such as data cleaning.
The dataset chosen was the Lyft Bike Sharing System in San Francisco Bay Area. With this data, we'll make several types of visualizations, such as univariate, bivariate, and multivariate plots. The main aim here is to explore many relationships in the data set.
The main goal of this investigation is to analyze the behavior of user types, see the paterns in bikes ride, periods with the most rides and the relation among all these variables. Beyond that a quick spatial view of the distribution of the bike stations.
The dataset was downloaded from Lift website. In the CSV file there are a lot of information, such as duration of the ride, when and where it started, the type of user and the name of stations, beyond the bike number and location (latitude and longitude). This dataset refers to the 2nd semester of 2017.
Bikes play a tremendous role in the population daily basis of transportation.
They're used to move between two transports modal, beyond being used in medium distances.
Subscribers ride are quicker than customers. This might be because subscribers use them daily with a pouporse: moving around the city in a quick and efficient way. While customers take longer rides. This might happen because they use once in a while, or for laisure porposes.
The peak hour for using the bikes are in the afternoon. And the stations with most bike returns are those ones with other transport connection or a croweded region.
The rides avarege is about 18min. But if we take a look for each user type, we'll see that customers ride for 17.5min and subscribers about 8min.