Abstract
dc:descriptionSentiment Analysis constitutes a key data analytics tool in many contexts and domains, since it helps to automatically detect and analyze public opinions, emotions, attitudes, and needs in massive amounts of unstructured data using Natural Language Processing and Text Mining methods. The research activity of this PhD thesis focused on two types of opinionated user-generated content; evaluations expressed by customers about products and services and their aspects in particular domains of interest (restaurant and laptop reviews), and verbal attacks against predefined target groups of interest (e.g. refugees, immigrants) in the context of Computational Social Sciences, covering an industrial and a humanitarian use case of Sentiment Analysis, respectively. In this setting, this thesis presents: a) a principled unified knowledge representation framework and English benchmark datasets for Aspect Based Sentiment Analysis, and b) a linguistically inspired and data-driven framework for examining Verbal Aggression as an indicator of xenophobic attitudes in Greek Social Media.
Degree
thesis:*- Grantor dc:publisher
- University of Crete (UOC)
- Year dc:date
- 2019
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Pontiki, Maria
- Ποντίκη, Μαρία
Subjects
dc:subject × 12Rights
dc:rights- Statement dc:rights
-
- BY_NC
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/46115