Doctoral Meeting: 'Corpus-based Construction of Polarity Lexicon'

Studies in sentiment analysis and opinion mining have been focused on many aspects related to opinions, namely polarity classification by making use of positive, negative or neutral values.

Sentiment lexicons are considered the primary building block in sentiment analysis as it is an essential resource for most sentiment analysis algorithms, and the first indicator to express positive or negative opinions.

This presentation will describe a corpus-based method to build a sentiment lexicons and how to use them to identify extreme opinions (most negative and most positive opinions).