Dark Mode
Daily Bicycle Share Demand and Seasonal Factors
Product Reviews & Feedback
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Predicting the fluctuating demand for shared bicycles involves analysing various environmental and temporal factors that influence rider behaviour. This collection of records provides the necessary metrics to explore how weather conditions, seasonal changes, and specific calendar dates impact the volume of bike rentals in an urban setting. By utilizing these variables, researchers can develop predictive models to estimate the total count of daily rentals, assisting in the optimisation of transport resources and the improvement of service availability.
Columns
- dteday: The specific date on which the bike rental was recorded.
- season: A categorical marker for the time of year, where 1 represents spring, 2 represents summer, 3 represents autumn, and 4 represents winter.
- yr: The year of the record, indicated as 0 for 2018 and 1 for 2019.
- mnth: The numerical month of the year, ranging from 1 to 12.
- holiday: A binary indicator showing whether the day was a public holiday (1) or not (0).
- weekday: The specific day of the week, indexed from 1 (Sunday) through to 7 (Saturday).
- workingday: A binary marker indicating if the day was a working day (1) or a weekend/holiday (0).
- weathersit: The weather condition at the time, ranging from 1 (clear or partly cloudy) to 4 (heavy rain, snow, or fog).
- temp: The recorded temperature in degrees Celsius.
- atemp: The "feels like" temperature in degrees Celsius, reflecting the perceived thermal comfort.
- hum: The recorded humidity percentage.
- windspeed: The speed of the wind at the time of the record.
- casual: The total count of rentals made by non-registered users.
- registered: The total count of rentals made by registered members of the service.
- cnt: The combined total of both casual and registered rentals for the day.
Distribution
The records are provided in a comma-separated values file titled
day.csv, with a total file size of 57.67 kB. The collection includes 750 valid records across 16 primary columns. Data integrity is high, with a 100% validity rate for core metrics, though 2 records are noted as missing in the detailed statistical summary. The resource maintains a usability score of 10.00 and is expected to receive quarterly updates.Usage
This resource is ideal for training multiple linear regression models to predict urban transport demand. It is well-suited for exploratory data analysis to identify the correlation between weather patterns and cycling habits. Furthermore, urban planners can use the daily rental counts to inform decisions regarding fleet management and the placement of new bicycle docking stations based on seasonal trends.
Coverage
The scope is focused on bike-sharing activities over a two-year period, specifically 2018 and 2019. It encompasses all four seasons and provides a granular look at how daily changes in weather and work schedules affect rental volume. The demographic scope is split between casual and registered users, offering a view of different customer segments.
License
CC0: Public Domain
Who Can Use It
Data science students can leverage these records to practice building and testing predictive algorithms for regression tasks. Urban transport analysts may utilise the figures to study how holidays and weather extremes impact alternative transit systems. Additionally, developers can use the structured date and weather fields to create data visualisations that track city-wide mobility patterns.
Dataset Name Suggestions
- Shared Bike Demand Prediction and Weather Metrics
- Urban Cycling Rental Log: 2018-2019
- Daily Bicycle Share Demand and Seasonal Factors
- Predictive Modelling for Shared Transport Volume
- Westeros Bike Share: Two-Year Usage Analytics
Attributes
Original Data Source: Daily Bicycle Share Demand and Seasonal Factors
QUALITYLoading...
Free
Download Dataset in CSV Format
Recommended Datasets
Loading recommendations...