Predicting Bike Availability Using The Linear Regression Machine Learning Algorithm
This project was submitted as my undergraduate computer science capstone project. Red Bike Adventures is a fictional company created by me for the purposes of this project. To visit the website created here, please go to: https://beth-culler.vercel.app/
The goal of this project is to solve the following business need:
"The Little Red Rock Trail is one of the most visited biking trails in Nevada, with roughly one million visitors each year, due to its fun ride and amazing views of the valley in Red Rock Canyon. Most visitors are able to enjoy cycling the Little Red Rock Trail by renting a bike from the Red Bike Adventures bike station located at the trailhead. Due to its popularity, many non-local visitors would like to try to plan their trip to the area during a time they can be more guaranteed a bike to rent, so many call inquiring about what times those may be. Due to the many factors that contribute to bike availability (temperature, time of year, local events, etc.), it is very difficult to provide customers with a reliable answer. Currently customers can only drive out to the rural area and hope a bike is still available to rent. This uncertainty may deter customers from visiting. Red Bike Adventures is a tourism-dependent company and thus needs to encourage as many visitors as possible.
In resolving this, the solution will provide customers with useful information regarding historical trends and future predictions of bike availability to reduce uncertainty and therefore encourage tourism to continue to thrive. This will ensure that Red Bike Adventures remains the number one provider of bike sharing systems in Nevada."
The dataset that was used is the public dataset from the University of California Irvine’s Machine Learning Repository which consists of a two-year historical log corresponding to the years 2011 and 2012 from Capital Bikeshare system, Washington D.C. The description of this dataset was changed to fit the narrative that was made up for this project. The public dataset can be found below:
https://archive-beta.ics.uci.edu/ml/datasets/275
Use of this dataset in publications is required to be cited to the following publication:
Fanaee-T, Hadi, and Gama, Joao, "Event labeling combining ensemble detectors and background knowledge", Progress in Artificial Intelligence (2013): pp. 1-15, Springer Berlin Heidelberg, doi:10.1007/s13748-013-0040-3.
@article{ year={2013}, issn={2192-6352}, journal={Progress in Artificial Intelligence}, doi={10.1007/s13748-013-0040-3}, title={Event labeling combining ensemble detectors and background knowledge}, url={http://dx.doi.org/10.1007/s13748-013-0040-3}, publisher={Springer Berlin Heidelberg}, keywords={Event labeling; Event detection; Ensemble learning; Background knowledge}, author={Fanaee-T, Hadi and Gama, Joao}, pages={1-15} }