Definition and scope of sentiment analysis

Summary

Sentiment analysis (also called opinion mining) is the use of natural language processing, text analysis, computational linguistics, and related techniques to systematically identify, extract, quantify, and study affective states and subjective information in text. Its scope includes analyzing opinions and emotions expressed in many kinds of materials—such as customer reviews, survey responses, social media, and healthcare-related documents—and it can support applications ranging from marketing and customer service to clinical medicine. In terms of what it covers, sentiment analysis can operate at multiple levels: document-level, sentence-level, and feature/aspect level. A core task is determining sentiment polarity (positive, negative, or neutral), while more advanced “beyond polarity” approaches classify specific emotions (e.g., anger, fear, sadness) and can also estimate sentiment intensity (how strong the sentiment is). The field also includes related subtasks such as subjectivity/objectivity identification, which distinguishes factual statements from opinionated ones, and aspect-based sentiment analysis, which targets sentiment toward particular entity attributes (e.g., food quality vs. service).

Key Points

• 1

  Sentiment analysis identifies and quantifies affective states and subjective information using NLP and text analysis.

• 2

  It spans multiple granularity levels (document, sentence, feature/aspect) and can go beyond polarity to emotions, intensity, and subjectivity/objectivity.

• 3

  Common scope includes tasks like aspect-based sentiment analysis and subjectivity detection, supporting applications such as customer feedback analysis and decision-making.