Analysis on CloudPBX's data. The data that was worked on was kindly provided by CloudPBX, they are a PBX provider who service both independent customers and businesses. They were interested in investigating how they could better improve their services so they took part in the BCData Science Workshop which is a week long workshop where participants will have the opportunity to work with big data, learn data analysis and so much more in a team setting: I highly recommend taking part in future events if you are given the chance.
More information can be found here.
This repo is a culmination of the work I completed during this workshop. I would definitely have to note that this is product of the collaboration between all my team members during the event (stefanhannie,imachuca24,mlardadji,eclam,jackiexu007 and clm222 and with the expertise of our supervisors (mattfung and grahamsnz). There is some work that is absent here, but if you are so inclined feel free to look at the master repo.
- Jupyter notebooks hosted by BCData
- Python 3.6
- Dask
- Pandas
- Numpy
- Matplotlib
- Geolite
- GeoIP2
- scipy
- xtensor
- Boost.python
- File Dump: there are some examples for coordinates and html pages used for the final mapping.
- Final Mapping: holds all the required files for the animation.
- Smaller Notebooks: holds notebooks on adding information to the data.
- Network Mapping: Data necessary to plot CouldPBX’s network.
- CloudPBX Workshop presentation: is the final presentation that the group published.
- Deep Dive into MOS: shows how MOS works and how it the data relates to it.
- MOS Qualfun Analysis: shows how Qualfun is calculated and how it highlights calls on the data.
- Call quality: analyses the quality of calls in the network.
- Loading 7 million entries shows how to load large datasets.