Automated road accident casuality survey based on news articles

The system is designed to gather information on road accident casualties by searching for and collecting news articles containing the keyword "road accident" from online news portals. The collected data is then processed through a natural language processing (NLP) model to extract information such as the number of deaths, injuries, location, and vehicle type involved in the accidents.

How to use

Install the dependencies

Install all required python libraries by running pip install -r requirements.txt It'll install the following requirements

• chromedriver_binary
• matplotlib==3.5.2
• nltk==3.7
• numpy==1.21.5
• pandas==1.4.3
• seaborn==0.11.2
• selenium==4.7.2
• word2number==1.1

Scrape news articles

Collect news articles from The Daily Star by running the following command python dailystar.py It'll open an automated google chrome window using chrome web driver and crawl through the website to collect news articles data and store them into input.csv file.

Run the NLP model

python model.py This command will pass the data from input.txt into the NLP model in order to extract the casuality data and put them into output.csv

Data Visualization

Use the jupyter notebook plots.ipynb to visualized the data in the output.csv file.