- Sentiment (opinion) - a view or attitude towards a situation or event
- Sentiment Analysis (also known as Opinion Mining or Emotion AI)
- refers to the use of natural language processing, text analysis, computational linguistics, and biometrics to systematically identify, extract, quantify, and study affective states and subjective information
- is a kind of Text Classification
- is the detection of attitudes given a body of Text:
- Attitude Source/Holder
- Attitude Target/Aspect/Attribute
- Attitude Type - from either:
- a set of types (e.g. like, love, hate, value, desire, etc)
- weighted polarity (e.g. positive-neutral-negative, together with strength)
Sentiment Analysis - Task Types
increasing complexity:
- is the attitude of text positive or negative
- rank the attitude of test from 1 to 5
- detect the target, source and/or attitude types
Sentiment Analysis - Classification Types
|
Class Type |
Description |
|---|---|
|
Polarity Classification |
basic task in sentiment analysis is classifying the polarity of a given text at the document, sentence, or feature/aspect level—whether the expressed opinion in a document, a sentence or an entity feature/aspect is:
|
|
Emotional Classification |
advanced, “beyond polarity” sentiment classification looks, for instance, at emotional states such as “angry”, “sad”, and “happy” |
|
Subjectivity/Objectivity Identification |
classifying a given text (usually a sentence) into one of two classes:
This problem can sometimes be more difficult than polarity classification |
|
Feature/Aspect-Based |
refers to determining the opinions or sentiments expressed on different features or aspects of entities different features can generate different sentiment responses (e.g. a hotel can have a convenient location, but mediocre food. location and food are different features of the entity hotel) |
Sentiment Analysis - Baseline Algorithm
see: Text Classification’s Baseline Algorithm section