Python中的情感分析：Textblob与Vader？

Question

I am learning how to conduct sentiment analysis on social media posts using Python. I found there are different tools to achieve this, such as Textblob or Vader.

I would like to understand what are the differences I should pay attention to when choosing one or the other.

In this case, I am trying to analyze tweets from different users over a fixed period of time.

Thanks!

Answer 1

Vader Sentiment Analysis works better for with texts from social media and in general as well. It is based on lexicons of sentiment-related words. Each words in the lexicon is rated whether it is positive or negative.

When it comes to analysing comments or text from social media, the sentiment of the sentence changes based on the emoticons. Vader takes this into account along with slang, capitalization etc and hence a better option when it comes to tweets analysis and their sentiments.

Refer to this paper for more details.

It produces 4 metrics - positive, negative, neutral and compound score. I did Twitter sentiment analysis using Vader and was surprised that the sentiments were better compared to textBlob. You can refer to the code in my githubrepo I manually validated random tweets and came up with a compound score that can determine positive and negative tweets.