Analyze and predict sentiment with machine learning

Sentiment analysis is the process of deriving the attitudes and opinions expressed in text data. It can be used to categorize subjective statements as positive, negative, or neutral in order to determine opinions or sentiment about a topic. Machine learning techniques are commonly used in sentiment analysis to build models that can predict sentiment in new pieces of text.

For example, sentiment about companies is often analyzed in the financial services industry, as public sentiment can impact financial markets. One might use the large amounts of historical news and past social media data to build a sentiment model using machine learning; this model could then be used to calculate in real time a public sentiment score from news and social media feeds indicating current public opinion.

Sentiment analysis is used in finance, software, Internet, and many other industries. Applications of sentiment analysis include:

  • Automating the classification of reviews, whether positive or negative
  • Analyzing surveys to understand why customers are satisfied or dissatisfied
  • Determining real-time sentiment scores for use in financial trading strategies
  • Assessing counterparty credit risk

To learn more about deriving understanding from sentiment and using machine learning, see Text Analytics Toolbox™ and Statistics and Machine Learning Toolbox™.

See also: data science, machine learning, deep learning, natural language processing , text mining, long short-term memory (LSTM) networks