Crawl Ask.fm QA lists and create corpus for ML.

This is a selenium tasks to crawl Ask.fm because of correcting QA list for Machine Learning.

• Background
• Install
• Usage
• Contributing
• License

Among machine learning, there was a task to create a bot that responds to natural language using LSTM (a kind of RNN).

At that time, a large amount of conversation corpus is required, but since I did not get a good conversation corpus, I decided to make a conversation corpus by crawling the Ask.fm question answer list with Selenium (Google Chrome) Did.

I'm using Selenium for Python because my favorite programming language is Python.

• Python 3.6+
• Google Chrome
• Google Chrome WebDriver
  • Check your Chrome version and install suitable driver version.

Install dependencies.

pip install -r requirements.txt

Before create conversation corpus, create face list because of crawling QA.

First args, number of loop count.

python src/get_faces.py 100

After run script, get face list into data/face_list.txt

python src/main.py

After run script, get conversation corpus into data/askfm_data/foobar.txt

See the contributing file!

PRs accepted.

Small note: If editing the Readme, please conform to the standard-readme specification.

MIT © tubone.