Opinion Mining and Analysis

*There are two main types in textual information

　　*Facts: Objective expressions about entities, events and their attributes; (like: i bought an iPhone yesterday)

　　*Opinions: Subjective expressions of sentiements, attitudes, emotions, appraisals or feeling toward entities, events, and their attributes (like: i really live this new iPhone)

*Opinions are important for people to make a decision: whenever people make a decision, they want to hear other's opinion.

*Opinion mining in the past: *individuals: opinions from friends and family; *Business: surveys, focus groups, consultants.

*Opinon mining in the present: *no limitation; *word-of-mouth on the web!

 

 

Main factors for Opinion mining:

　　*Target Object: an entity that can be a product, person, event, organisation, or topic(e.g. iPhone).

　　*Attribute: an object usually has two types of attributes

　　　　　　*Components: (e.g. touch screen, battery)

　　　　　　*Properties: (e.g. size, weight, colour, voice quality)

　　*Explicit and implicit attributes:

　　　　　　*Explicit attributes: appearing in the opinion (the battery life of this phone was not long)

　　　　　　*Implicit attributes: not appearing in the opinion (this phone is too expensive)

　　*Opinion Holder: the person or organisation that expresses the opinion (e.g. my mother was mad with me)

　　*Opinion orientation (polarity): positive, negative, or neutral.

==>a person or organisation that expresses a positive or negative sentiment on a particular attribute of an target object at a certain time.

 

*Dictionary-based Opinion Lexicion:

　　*Limitation: impossible to identify context-dependent opinion lexicons. (small:the screen is too small to read the documents; long:it takes a long time to set up)

 

*Opinion Mining Approaches:

　　*Document-level Opinion Mining:

　　*Sentence-level Opinion Mining:

　　*Rule-based Opinion Mining:

holder