Five-Dimensional Sentiment Analysis of Corpora, Documents and Words

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Researchers

Research units

  • University of Helsinki
  • National Consumer Research Centre

Abstract

Sentiment analysis has become a widely used approach to assess the emotional content of written documents such as customer feedback. In positive psychology research, the typical one-dimensional analysis framework has been extended to include five dimensions. This five-dimensional model, PERMA, enables a fine-grained analysis of written texts. We propose an approach in which this model, statistical analysis and the self-organizing map are used. We analyze corpora from various genres. A hybrid methodology that uses the self-organizing maps algorithm and human judgment is suggested for expanding the PERMA lexicon. This vocabulary expansion can be useful for English but it is potentially even more crucial in the case of other languages for which the lexicon is not readily available. The challenges and solutions related to the text mining of texts written in a morphologically complex language such as Finnish are also considered.

Details

Original languageEnglish
Title of host publicationAdvances in Self-Organizing Maps and Learning Vector Quantization - Proceedings of the 10th International Workshop, WSOM 2014
Publication statusPublished - 2014
MoE publication typeA4 Article in a conference publication
EventWorkshop on Self-Organizing Maps - Mittweida, Germany
Duration: 2 Jul 20144 Jul 2014
Conference number: 10

Publication series

NameAdvances in Intelligent Systems and Computing
Volume295
ISSN (Print)21945357

Workshop

WorkshopWorkshop on Self-Organizing Maps
Abbreviated titleWSOM
CountryGermany
CityMittweida
Period02/07/201404/07/2014

    Research areas

  • education, independent component analysis, life-philosophical lecturing, natural language processing, positive psychology, self-organizing map, Text mining

Download statistics

No data available

ID: 12864045