ABSTRACT
Academic literature has highlighted increasing interest in and value of sentiment analysis in academia and practice. Sentiment analysis remains a hotbed of research and development for academia and industry. This chapter seeks to provide a general introduction to the field and aims at highlighting some of the challenges that remain in advancing current approaches. Sentiment intensity shows sentiment orientations have different levels of strength, which can be identified through sentiment words with varied strength or words indicating intensifiers and diminishers. The use of intensifiers increases the degree of positivity or negativity, whereas using a diminisher decreases the degree of intensity. There are essentially two approaches to sentiment analysis: lexicon-based approaches and machine-learning approaches. There are three main existing approaches to compiling a list of sentiment words or a sentiment lexicon: manual approach, lexical approach and corpus-based approach. Aspect-level analysis is limited in identifying consumer sentiment expressed in a highly networked social media space.
