The purpose of this project is to visualize the document trail scraped from the Federal Energy Regulatory Commission's (FERC) library.
This project is installed along with FERC_DOC_TRAIL repository - it uses FERC_DOC_TRAIL's output and relies on full execution of all steps of FERC_DOC_TRAIL.
Clone the repository the regular way of cd'ing into a directory of choice and issuing the regular git clone command. This repository has to be cloned to the same directory as FERC_DOC_TRAIL, as seen on the picture:
cd Users/username
git clone https://github.com/ilyaperepelitsa/FERC_DOC_VIZ.git
cd FERC_DOC_VIZ
This visualization and analysis project relies on several Python libraries:
- Matplotlib for graphing data. Type the following to install:
python3 -mpip install matplotlib
- Pandas for tabular data analysis. Type the following to install:
pip3 install pandas
- Numpy for basic scientific computing. Type the following to install:
pip3 install matplotlib
This project has been optimized to filter and analyze the data gathered for CP16-17 docket. It replaces certain actors' names (there are typos and alternative ways of writing that were found for some actors, i.e. LLC/L.L.C. etc.) and adjusts other minor details. We will provide instructions on how to modify this file in the future, however currently it satisfies only the needs of monitoring CP16-17. You have already navigated into the directory in the previous section.
To run the script you need to issue the following command in the terminal:
python3 visualize.py
The script creates a folder called exhibits within itself. It saves two graphs in this directory and creates folders for tables. Currently tables include pivot tables of total counts of document accessions by two fields:
- Author Organization
- Recipient Organization
It groups the observations by organizations and a date which is represented in following formats:
- Month-Year
- Year
- Month (to explore potential seasonal trends, i.e. whether there have been more accessions towards the end of the year)