View Item 
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        JavaScript is disabled for your browser. Some features of this site may not work without it.

        Browse

        All of UU Student Theses RepositoryBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

        News for the Advanced: A Simple Approach to Automatically Detecting Satire

        Thumbnail
        View/Open
        tenWolde_6279945.pdf (421.4Kb)
        Publication date
        2021
        Author
        Wolde, M.H. ten
        Metadata
        Show full item record
        Summary
        Figurative language is seen as the topic that pains Natural Language Processing (NLP) the most. The intention of the figurative language goes beyond the literal meaning of the words. This makes it a real challenge for both humans and computers. Automatic implementations for irony and sarcasm detection are widely researched in English. Satire is a form of figurative language that tries to imitate real news, exposing individuals or organizations to ridicule. Earlier work suggests that an automated assistive tool that detects satire could be the first step in fighting the growing world of fake news. This paper introduces a simple machine learning classifier to automatically detect satire in Dutch headlines, using general features based on textual markers and sentiment. The results of our classifier are promising for further research on automatic figurative language detection in Dutch, but there is some work to do there. Some suggestions for future research are also included
        URI
        https://studenttheses.uu.nl/handle/20.500.12932/39518
        Collections
        • Theses
        Utrecht university logo