Analyze and visualize ride-sharing data using the power of Python, Pandas, and Matplotlib
Analyzing company profits and customer demographics for a ride sharing app:
- A bubble chart that showcases the average fare versus the total number of rides with bubble size based on the total number of drivers for each city type
- The mean, median, and mode for the following:
- The total number of rides for each city type.
- The average fares for each city type.
- The total number of drivers for each city type.
- Box-and-whisker plots that visualize each of the following to determine if there are any outliers:
- The number of rides for each city type.
- The fares for each city type.
- The number of drivers for each city type.
- A pie chart that visualizes each of the following data for each city type:
- The percent of total fares.
- The percent of total rides.
- The percent of total drivers.
Data Sources: city_data.csv ride_data.csv
Software: Python 3.7.1 Pandas 0.25.2 Matplotlib 3.1.1
See PyBer.ipynb for summary statistics
Providing a summary table of key metrics. Including, ride-sharing data / city type, and a multi-line graph indicating average fare / week / city type.
Findings show great amounts of business conducted in urban cities, in comparison to suburban or rural (4-10x number of rides, & 5-30x number of drivers). Furthermore, the average fares are the lowest for drivers and rides in urban cities (indicating shorter duration, & cheaper trips in urban areas - as well as the high volume of drivers, & less earned $ on average in comparison to rural and/or suburban data).
Showing trends across city types: Urban, outrighly accounting for the most expensive fares, followed by suburban, and lastly rural. The city types do appear to mirror eachother with perhaps a passive relationship (rising and plateauing in similar fashion Jan - Apr).
Indicates a solid visual represenation of what may be socioeconomic bracketing / city type.
While there are no intersectional point between the difference city types, their trend lines move in unison through time - giving us a vague interpretation, but an insight nonetheless (that we could absolutely look further into if prompted), to macro-economic conditions & factors which may play a role in market volatility (fare$).
