Utilize este identificador para referenciar este registo:
https://hdl.handle.net/1822/65969
Título: | Predicting performance problems through emotional analysis |
Autor(es): | Martins, Ricardo Almeida, José João Henriques, Pedro Rangel Novais, Paulo |
Palavras-chave: | Emotion analysis Machine learning Natural processing language |
Data: | 2018 |
Editora: | Schloss Dagstuhl – Leibniz-Zentrum für Informatik GmbH |
Revista: | OASIcs: OpenAccess Series in Informatics |
Resumo(s): | In the cartoons, every time a character is nervous he/she begins to count to ten to keep calm. This is a technique, among hundreds, that helps to control the emotional state. However, what would be the impact if the emotions would not be controlled? Are the emotions important in terms of impairing the ability to perform tasks correctly? Using a case study of typing text, this paper is about a process to predict the number of writing errors from a person based on the emotional state and some characteristics of the writing process. Using preprocessing techniques, lexicon-based approaches and machine learning, we achieved a percentage of 80% of correct values, when considering the emotional profile on the writing style. |
Tipo: | Artigo em ata de conferência |
URI: | https://hdl.handle.net/1822/65969 |
ISBN: | 9783959770729 |
DOI: | 10.4230/OASIcs.SLATE.2018.19 |
ISSN: | 2190-6807 |
Versão da editora: | https://drops.dagstuhl.de/opus/volltexte/2018/9277/pdf/OASIcs-SLATE-2018-19.pdf |
Arbitragem científica: | yes |
Acesso: | Acesso aberto |
Aparece nas coleções: |
Ficheiros deste registo:
Ficheiro | Descrição | Tamanho | Formato | |
---|---|---|---|---|
OASIcs-SLATE-2018-19.pdf | 427,09 kB | Adobe PDF | Ver/Abrir |
Este trabalho está licenciado sob uma Licença Creative Commons