Artificial immune systems for sentiment analysis
Τεχνητά ανοσοποιητικά συστήματα για την ανάλυση συναισθήματος
View/
Keywords
Artificial immune systems ; Artificial Immune Recognition Systems (AIRS) ; Sentiment analysis ; Artificial intelligence ; Machine learning ; ClassificationAbstract
This thesis aims to present AIRS-x, an alternative formulation to the Artificial Immune System (AIRS)
algorithm. AIRS is a supervised learning classification algorithm, which has emerged from the field of
Artificial Immune Systems (AIS) – or algorithms inspired by immunological concepts. We present a broad
overview of machine learning and immunology; we describe how AIS emerged as a computational
paradigm based on biological metaphors; and we explore the scope and the wide range of applications of
this relatively novel field. Furthermore, we introduce the AIRS algorithm, describing its workflow, its
advantages, and its limitations. In addition, the thesis showcases a novel formulation submitted by
Giatzitzoglou (2018) and Giatzitzoglou et al. (2019), the AIRS-x algorithm, indicates how it enhances the
original, explores its newly introduced parameters and its promise as a sentiment analysis tool. Finally, it
suggests paths for future exploration.
Postgraduate Studies Programme
ΠληροφορικήDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα ΠληροφορικήςNumber of pages
68Language
EnglishCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα