Acquire, analyze, and summarize data from all corners of the social web.

The repository has three parts:

1. Scraper, to obtain data from social media.

2. Training script, to train a classifier.

3. A script to classify text using the previous trained classifier.

To use the classifier, the following libraries must be installed:

i. scikit-learn==0.18.2

ii. nltk==3.2.4

iii. numpy==1.13.1

iv. pandas==0.20.3

v. scipy==0.19.1

vi. twitter==1.17.1

Download Docker and build the image as follows:

docker build -t mining_the_social_web ~/Mining_The_Social_Web/

then, run the container with the instruction:

docker run --rm -it -v $(pwd):/Mining_The_Social_Web mining_the_social_web

and it should be ready to use.