csr_analysis

This repository makes word clowd from Japanese company's CSR report.

The application consists of 2 steps.

1. PDF to text
2. Make word cloud from text file

Installation

You have to use python3.

Install package from requirements.

pip install -r requirements.txt

Usage

PDF to text

Make text file from CSR report with PDFMiner.

First, you ｐlaced PDF File into the raw directory. Next, edit to parameters you want.

input_file = './data/raw/XXX.pdf'
interim_dir = './data/interim/XXX/'
processed_file = './data/processed/XXX.txt'

Text file is generated after executing the below command.

python pdf_to_text.py

Make word cloud from text file

Making word cloud on iPython notebook.

ipython notebook

Open wordcloud.ipynb and edit the parameter of the first cell to the file name made from previous step.
file_name = './data/processed/XXX.txt'

After that, run all cells.