This is part of the code used within my Third Year Project at the University Of Southampton. This, combined with other elements, formed the basis of my dissertation, of which the IP belongs to the University Of Southampton (as is default).
The majority of the code stored here retrieves and processes the data available through the National Rail Open Data Feeds.
This repository includes:
- Set up of database required to hold data.
- Data Characterisation scripts examining what data was collected and it's general characteristics
- Scripts experimenting with the various data feeds (particularly the Open Data Feeds and Darwin Feed)
- Python scripts to preprocess the data before machine learning takes place (data cleaning, sanitisation etc)
- Python scripts using sk-learn among other libraries to experiment with various models and processes to learn from the data (note this takes place within the 'Preprocessing.py' script). This was the experimentation and training phase.
- Python scripts denoting the final method used to manipulate the data to give a model and results.
- Graphing scripts