Authors:
(1) Florian Cafiero (ORCID 0000-0002-1951-6942), Sciences Po, Medialab;
(2) Jean-Baptiste Camps (ORCID 0000-0003-0385-7037), Ecole nationale des chartes, Universite Paris, Sciences & Lettres. Table of Links Abstract and Introduction Why work on QAnon? Specificities and social impact Who is Q? The theories put to test Authorship attribution Results Discussion Corpus constitution Quotes of authors outside of the corpus have been Definition of two subcorpus: dealing with generic difference and an imbalanced dataset The genre of “Q drops”: a methodological challenge Detecting style changes: rolling stylometry Ethical statement, Acknowledgements, and References The genre of “Q drops”: a methodological challenge The study of the Q drops raises a number of specific challenges. First of all, the Q drops constitute per se a kind of a genre. It follows specific rules that most people would not use in another context: they do not look like a regular blog, media or social media post, nor do they belong to any specific literary genre etc. This forces us to consider our attribution problem as a cross-topic attribution problem. This specific genre has consequences on many linguistic properties of the Q drops. The structural brevity of the sentences for instance prevents us from taking the sentence lengths as a clue of who wrote what. The overwhelming proportion of interrogative forms, especially in the first Q drops, makes it difficult to reason on morpho-syntactic sequences, as they are often extremely and artificially stereotyped. Part-of-Speech n-grams such as “Interrogative pronoun - conjugated verb - common noun” would for instance emerge as a signature of the Q drops, and could derail our analysis, by pointing to any of the suspects using interrogative forms the most in other contexts. Finally, the elliptic style of the Q drops, written almost as if they were a telegram, distorts the use of function words, less frequent than expected in the Q drops corpus. Approaches relying only on function words could be made less robust by this distortion. We thus chose to work on character trigrams, the most flexible and reliable feature we could use in the very specific context of this study, and a widely acknowledge feature in stylometry, in particulary for its supposed capacity to capture grammatical morphemes (Kestemont, 2014; Sapkota et al., 2015) while bearing in mind potential greater sensibility to thematic attractions in comparison to function words. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Florian Cafiero (ORCID 0000-0002-1951-6942), Sciences Po, Medialab; (2) Jean-Baptiste Camps (ORCID 0000-0003-0385-7037), Ecole nationale des chartes, Universite Paris, Sciences & Lettres. Authors: Authors: (1) Florian Cafiero (ORCID 0000-0002-1951-6942), Sciences Po, Medialab; (2) Jean-Baptiste Camps (ORCID 0000-0003-0385-7037), Ecole nationale des chartes, Universite Paris, Sciences & Lettres. Table of Links Abstract and Introduction Abstract and Introduction Why work on QAnon? Specificities and social impact Why work on QAnon? Specificities and social impact Who is Q? The theories put to test Who is Q? The theories put to test Authorship attribution Authorship attribution Results Results Discussion Discussion Corpus constitution Corpus constitution Quotes of authors outside of the corpus have been Quotes of authors outside of the corpus have been Definition of two subcorpus: dealing with generic difference and an imbalanced dataset Definition of two subcorpus: dealing with generic difference and an imbalanced dataset The genre of “Q drops”: a methodological challenge The genre of “Q drops”: a methodological challenge Detecting style changes: rolling stylometry Detecting style changes: rolling stylometry Ethical statement, Acknowledgements, and References Ethical statement, Acknowledgements, and References The genre of “Q drops”: a methodological challenge The study of the Q drops raises a number of specific challenges. First of all, the Q drops constitute per se a kind of a genre. It follows specific rules that most people would not use in another context: they do not look like a regular blog, media or social media post, nor do they belong to any specific literary genre etc. This forces us to consider our attribution problem as a cross-topic attribution problem. This specific genre has consequences on many linguistic properties of the Q drops. The structural brevity of the sentences for instance prevents us from taking the sentence lengths as a clue of who wrote what. The overwhelming proportion of interrogative forms, especially in the first Q drops, makes it difficult to reason on morpho-syntactic sequences, as they are often extremely and artificially stereotyped. Part-of-Speech n-grams such as “Interrogative pronoun - conjugated verb - common noun” would for instance emerge as a signature of the Q drops, and could derail our analysis, by pointing to any of the suspects using interrogative forms the most in other contexts. Finally, the elliptic style of the Q drops, written almost as if they were a telegram, distorts the use of function words, less frequent than expected in the Q drops corpus. Approaches relying only on function words could be made less robust by this distortion. We thus chose to work on character trigrams, the most flexible and reliable feature we could use in the very specific context of this study, and a widely acknowledge feature in stylometry, in particulary for its supposed capacity to capture grammatical morphemes (Kestemont, 2014; Sapkota et al., 2015) while bearing in mind potential greater sensibility to thematic attractions in comparison to function words. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Part of HackerNoon's growing list of open-source research papers, promoting free access to academic material.

Analyzing QDrops Texts shows Distinct Linguistic Features than Traditional Ones

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Machine Learning Reveals the Faces Behind Viral QAnon Posts

A Mathematical Overview of Dimension Reduction in Text Classification

A Brief Introduction to Statistical Parsing

When Words Won’t Talk, Sentence Structures Spill the Truth

Can AI Tell Jane Austen’s Writing Apart from a Fake?

Deep Syntax and Dead Founders: How AI Deciphered The Federalist Papers

Machine Learning Reveals the Faces Behind Viral QAnon Posts

A Mathematical Overview of Dimension Reduction in Text Classification

A Brief Introduction to Statistical Parsing

When Words Won’t Talk, Sentence Structures Spill the Truth

Can AI Tell Jane Austen’s Writing Apart from a Fake?

Deep Syntax and Dead Founders: How AI Deciphered The Federalist Papers

Light-Mode

Classic

Newspaper

Minty

Dark-Mode

Neon Noir

Minty

HN StartUps