Springboard Data Science Career Track: The second capstone project. The purpose of this project was to expand my knowledge in data science and make something useful while learning the data science stack. The predictive model used was a support vector regression and the experiment has been deemed a success by successfully predicting the trend of the test data with a 90% or higher P-value.
The directory "data" is purposefully ignored because the data set is too large for git hub. The data can be found at data.world:
Data sources can be viewed in the "sources/reference.md" file.
Detailed Analysis and working Python code can be found in the "PredictingFlightDelays.ipynb" Jupyter Notebook.
SVM parameters and metrics can be found in the "reports/model_metrics.csv" table.
/ - Root directory contains all project files and hosts the "PredictingFlightDelays.ipynb" Jupyter Notebook and this readme.md file.
|/data - Houses all .csv files from raw and processed data. (ignored)
|/figures - Contains all plots and charts that have been generated and modified.
|/reports - Contains bottom line reports and articles about the project.
|/sandbox - Experiments and code snippets that did not make it into the final Jupyter Notebook. (ignored)
|/sources - Project directions, links to data sources, and references used in the report.