Sentiment and emotion analysis from textual information: A systematic literature review

The aim of this article is to compile and expose the different methods of analysis of emotions and feelings from textual information. A Systematic Literature Review (SLR) was conducted to answer the following questions: What are the methods of analysis of emotions and feelings based on a text that predominates in academic literature? A narrative analysis approach was applied in order to comprehensively understand, locate and synthesise the studies related to the methods used in the analysis of feeling and emotion. The selection of the sample articles to be reviewed was made by means of inclusion and exclusion criteria. The results obtained show that there are several methodologies for sentiment and emotion analysis from digital textual information. Among these, Machine Learning (ML) emerges as particularly well-suited for analysing unstructured data, outperforming other approaches in its ability to manage large data sets and uncover complex patterns. Its application, both in supervised and unsupervised frameworks, has proven highly effective in integrating psychological constructs into textual analysis.