Sentiment Analysis

Sentiment analysis is a Natural Language Processing (NLP) technique to analyze (identify, extract, quantify) the sentiment(the opinion, judgment or emotion) in a piece of text.
ℹ Sentiments are measured level of polarity by either labels(positive, neutral, and negative, Compound) or by a value.


Applications:

  • Social media monitoring(Brand Monitoring, Or Brand Reputation Analysis)
  • Analyzing customer texts & voices
  • Enhancing client support analysis
  • Employee monitoring
  • Customer Feedback
  • Campaign Monitoring
  • Stock Market Analysis
  • Compliance Monitoring
  • Market Research Survey

Approaches:


Related topics:

  • Latent Semantic Analysis(LSA)

Algorithms:

  • Naive Bayes
  • VADER(Valence Aware Dictionary for Sentiment Reasoning) model

Notes:

  • Sentiments are often measured using positive, negative, or neutral categorical variables.
  • Depending on the application of Sentiment Analysis tool, it's used on social media posts, blog posts, reviews, surveys, or publications.
  • Depending on the application, this tool can be used beside Named Entity Recognition (NER) to identify the target of the sentiment.

References: