Misinformation on Twitter During the Danish National Election: A Case Study

Leon Derczynski, Torben Oskar Albert-Lindqvist, Marius Venø Bendsen, Nanna Inie, Viktor Due Pedersen and Jens Egholm Pedersen

TRUTH & TRUST ONLINE 2019

Elections are a time when communication is important in democracies, including over social media. This paper describes a case study of applying NLP to determine the extent to which misinformation and external manipulation were present on Twitter during a national election. We use three methods to detect the spread of misinformation: analysing unusual spatial and temporal behaviours; detecting known false claims and using these to estimate the total prevalence; and detecting amplifiers through language use. We find that while present, detectable spread of misinformation on Twitter was remarkably low during the election period in Denmark.

pdf

Political Stance Detection for Danish

Rasmus Lehmann and Leon Derczynski

NODALIDA 2019

The task of stance detection consists of classifying the opinion expressed within a text towards some target. This paper presents a dataset of quotes from Danish politicians, labelled for stance, and also stance detection results in this context. Two deep learning-based models are designed, implemented and optimized for political stance detection. The simplest model design, applying no conditionality, and word embeddings averaged across quotes, yields the strongest results. Furthermore, it was found that inclusion of the quote’s utterer and the party affiliation of the quoted politician, greatly improved performance of the strongest model.

Dansk abstrakt: I indeværende artikel præsenteres et annoteret datasæt over citater fra danske politikere, samt to Deep Learning-baserede modeller til brug ved identifikation af holdninger i de annoterede citater. Det konkluderes at den simpleste af de to modeller opnår de bedste resultater, samt at brug af information vedrørende citaternes kontekst forbedrer modellernes resultater.

pdf

Bornholmsk Natural Language Processing: Resources and Tools

Leon Derczynski and Alex Speed Kjeldsen

NODALIDA 2019

This paper introduces language processing resources and tools for Bornholmsk, a language spoken on the island of Bornholm, with roots in Danish and closely related to Scanian. This presents an overview of the language and available data, and the first NLP models for this living, minority Nordic language.

Sammenfattnijng pa borrijnholmst: Dæjnna artikkelijn introduserer natursprågsresurser å varktoi for borrijnholmst, ed språg a dær snakkes på ön Borrijnholm me rødder i danst å i nær familia me skånst. Artikkelijn gjer ed âuersyn âuer språged å di datan som fijnnes, å di fosste NLP modællarna for dætta læwenes nordiska minnretâlsspråaged.

pdf

The Lacunae of Danish Natural Language Processing

Andreas Kirkedal, Barbara Plank, Leon Derczynski and Natalie Schluter

NODALIDA 2019

Danish is a North Germanic language spoken principally in Denmark, a country with a long tradition of technological and scientific innovation. However, the language has received relatively little attention from a technological perspective. In this paper, we review Natural Language Processing (NLP) research, digital resources and tools which have been developed for Danish. We find that availability of models and tools is limited, which calls for work that lifts Danish NLP a step closer to the privileged languages.

Dansk abstrakt: Dansk er et nordgermansk sprog, talt primært i kongeriget Danmark, et land med stærk tradition for teknologisk og videnskabelig innovation. Det danske sprog har imidlertid været genstand for relativt begrænset opmærksomhed, teknologisk set. I denne artikel gennemgar vi sprogteknologi-forskning, -ressourcer og -værktøjer udviklet for dansk. Vi konkluderer at der eksisterer et fatal af modeller og værktøjer, hvilket indbyder til forskning som løfter dansk sprogteknologi i niveau med mere priviligerede sprog.

pdf

UniParse: A universal graph-based parsing toolkit

Daniel Varab and Natalie Schluter

NODALIDA 2019

This paper describes the design and use of the graph-based parsing framework and toolkit UniParse, released as an open-source python software package. UniParse as a framework novelly streamlines research prototyping, development and evaluation of graph-based dependency parsing architectures. UniParse does this by enabling highly efficient, sufficiently independent, easily readable, and easily extensible implementations for all dependency parser components. We distribute the toolkit with ready-made configurations as reimplementations of all current state-of-the-art first-order graph-based parsers, including even more efficient Cython implementations of both encoders and decoders, as well as the required specialised loss functions.

pdf