This dataset provides a group of information about rides in company called Ford GoBike which do their activities around San Francisco Bay Area.The company business model based on renting bikes , for instance you can take a bike from one of the stations which are Prevalent throughout the whole city then give it back after finishing your ride somewhere else within the city's stations.When it comes to payment they have a several options for clients like , annual subscription ,purchasing 3 days or 1 day (24 hours).
Finding out the key insights and the patterns and trends in clients' usage to help the company improve its performance.
In the exploration,We found that being in the "bike share for all" program doesn't impact the ride duration;because both of them almost had the same average.As we might expect users spent more time on biking at weekends where the trip duration reached its peak in weekends by about 15mins. Moreover,the average amount of time spent for biking for men is slightly less than those for women. Previously, we knew that almost 75% of our data are men, which indicates that men is more determined with the period they'll spend in biking.
Outside of the main variables of interest, just Out of curiosity I want to know when did users use bike for longer times? ,I found that 3 AM had the maximum average ride duration by more than 25 mins,then 2 Am came in the second place by 17 mins.Furthermore,The vast majority of subscribers are men.
For the presentation I focused on the positive correlation between increasing the use of bikes and being in workdays of the week and more specifically the hours during the day .I wanted also to point to rush hours for bike-share usage in terms of hours in the day and days in the week. Accordingly, both of them assured the correlation we mentioned.Then, we looked at the different ternds between subscribers and customers.Eventually, I looked at the relationship between [age or gender] and user type and how that may impact the trip duration.I've made sure to use different color palettes for each quality variable to make sure it is clear that they're different between plots.
- Feature engineering
- Data Visualization:
- Univariate Exploration
- Bivariate Exploration
- Multivariate Exploration
- etc.
- Python
- Pandas, , Numpy
- Matplotlib, Seaborn
- jupyter
- etc.
Recources : https://matplotlib.org/3.2.1/api/_as_gen/matplotlib.pyplot.subplot.html https://blog.datawrapper.de/beautifulcolors/#:~:text=Use%20warm%20colors%20&%20blue%20There%E2%80%99s%20a%20complementary,loved%20by%20data%20visualization%20designers:%20yellow/orange/red%20and%20blue