YouTube-Sentiment-Analysis

Scrape all the YouTube comments using api.

Description :

This project works by scraping YouTube comments and identify the sentiment of comments.

The directory FancySentiment shows the WordCloud (most frequent words) of the comments.

The directory CommentSentiment shows the positive/negative sentiment (using NaiveBayesClassifier) of the comments.

Prerequisite :

Python 3

pip(Python Package Index) :

$ sudo apt-get install python3-pip

requests package :

$ sudo pip3 install requests

lxml package :

$ sudo apt-get install libxml2-dev libxslt1-dev python-dev

$ pip3 install lxml

matplotlib package :

$ sudo pip3 install matplotlib

nltk package :

$ sudo pip3 install nltk

wordcloud package :

$ sudo pip3 install wordcloud

Running the file :

First input the API key in the file.

CommentSentiment/comment_extract.py

FancySentiment/comment_downloader.py

Then run the driver.py file.