ORCID
- Palomino, Marco: 0000-0001-7850-416X
Abstract
Sentiment analysis, the classification of human emotion expressed in text, has the potential to enhance our ability to analyse the ever growing amount of information published each day on social media. Thus, we compare here seven of the most well-regarded sentiment analysis tools, and conclude that none of them is sufficiently reliable to be used on its own. Combining them and relying on their results only when various tools reach an agreement seems to be a better option. The pros and cons of such an approach are discussed in this paper, while providing recommendations related to the usability of the tools in question. Our work is of particular relevance to small and medium-sized enterprises (SMEs), which constitute a large and integral part of the economy. SMEs seem to be ideal candidates to turn data derived from sentiment analysis into business opportunities.
DOI
10.1007/978-3-030-66527-2_5
Publication Date
2020-12-31
Publication Title
Lecture Notes in Computer Science
ISSN
0302-9743
Embargo Period
2021-03-23
Organisational Unit
School of Engineering, Computing and Mathematics
Recommended Citation
Palomino, M., Varma, A., & Connelly, A. (2020) 'Investigating the Lack of Consensus Among Sentiment Analysis Tools', Lecture Notes in Computer Science, . Available at: https://doi.org/10.1007/978-3-030-66527-2_5
