No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Verbal Aggression as an Indicator of Xenophobic Attitudes in Greek Twitter during and after the Financial Crisis
... The increasing number of users connecting through social networks and web platforms, such as Facebook and Twitter, as well as numerous Blogs and Wikis, creates continuously a significant volume in written communication through the Web [1][2][3][4][5][6][7]. The amount and quality of information and knowledge extracted from social text has been considered crucial to studying and analyzing public opinion [1,3,5,8,9], as well as linguistic [2,7,[10][11][12][13][14][15] and behavioral [4,6,[16][17][18] patterns. In its typical form, social text is often short in length, low in readability scores, informal, syntactically unstructured, characterized by great morphological diversity and features of oral speech, misspellings and slang vocabulary, consequently presenting major challenges for NLP and Data Mining tasks [2,4,7,10,11,13,[13][14][15][16]19]. ...
... Baxevanakis et al. [11] created a novel, manually annotated, social text corpus 18 written in Modern Greek and presented an NLP framework for gender identification of the 18 Corpus and code can be provided upon request. ...
... There are several approaches that have attempted to detect and analyze bullying and aggressive behavior in Virtual Learning Communities (VLCs) [4,16,17]. Other work focuses on offensive language identification and analysis in tweets [6,18]. An overview of the recent literature regarding offensive behavior and language detection, which is discussed in this subsection, is shown in Tables 3 and 4. ...
Mining social web text has been at the heart of the Natural Language Processing and Data Mining research community in the last 15 years. Though most of the reported work is on widely spoken languages, such as English, the significance of approaches that deal with less commonly spoken languages, such as Greek, is evident for reasons of preserving and documenting minority languages, cultural and ethnic diversity, and identifying intercultural similarities and differences. The present work aims at identifying, documenting and comparing social text data sets, as well as mining techniques and applications on social web text that target Modern Greek, focusing on the arising challenges and the potential for future research in the specific less widely spoken language.
... Theme 1: Antisemitic language correlates with the timing of key social, political, and historical events, many of which are societal failures that have no apparent connection to Jewish people. In a study on xenophobia in Greece, Pontiki et al. (2020) found that attacks on Jewish people increased with the rise of the far-right Golden Dawn party-which normalized antisemitic attitudesand with Greece's financial crisis-which led to economic conspiracies of Jewish people being singled out as the group to blame for the world's financial troubles. In Hungary, the Kuruc.info ...
... According to Ozalp et al. (2020), tweets from organizations combatting antisemitism gained more traction than antisemitic tweets, suggesting that "collective efficacy"-the ability of members of a community to control behavior within the said community-could be powerful. Pontiki et al. (2020) further corroborate this suggestion with their finding that antisemitic attacks decreased when Greece's Golden Dawn party was labeled as a criminal party. Comerford and Gerster (2021) recommend that social media platforms address antisemitism as part of a larger digital regulation initiative that includes education about common forms of antisemitism. ...
Automatic hate speech detection is an important yet complex task, requiring knowledge of common sense, stereotypes of protected groups, and histories of discrimination, each of which may constantly evolve. In this paper, we propose a group-specific approach to NLP for online hate speech detection. The approach consists of creating and infusing historical and linguistic knowledge about a particular protected group into hate speech detection models, analyzing historical data about discrimination against a protected group to better predict spikes in hate speech against that group, and critically evaluating hate speech detection models through lenses of intersectionality and ethics. We demonstrate this approach through a case study on NLP for detection of antisemitic hate speech. The case study synthesizes the current English-language literature on NLP for antisemitism detection, introduces a novel knowledge graph of antisemitic history and language from the 20th century to the present, infuses information from the knowledge graph into a set of tweets over Logistic Regression and uncased DistilBERT baselines, and suggests that incorporating context from the knowledge graph can help models pick up subtle stereotypes.
... As a result, Albanians in Greece have been stigmatized as "cunning", "primitive", "untrustworthy", "dangerous" and "criminal" (Lazaridis and Wickens 1999, p. 648) and also have been perceived as "polluters" and "intruders" of the "pure" and "homogenous" Greek identity (Psimmenos 2001, p. 32). Although the current discourses in Greece evolve around the successful integration of Albanians and claim that their stigmatization is part of the early reception of Albanians in Greece, recent research shows that these discourses are very much present today (Pontiki et al. 2020;Ndoci 2021a). Presently, references to Albanians are occasionally blatantly xenophobic as in the case of Greek tweets (Pontiki et al. 2020) or a January 2022 comment by Nikos Evangelatos, one of the major news anchors of Greek television, who added Tι πιo σύνηθες απóέναν Aλβανó µε óπλo; "What is more common than an Albanian carrying a gun?" during a news report about a recent arrest of an Albanian man. ...
... Although the current discourses in Greece evolve around the successful integration of Albanians and claim that their stigmatization is part of the early reception of Albanians in Greece, recent research shows that these discourses are very much present today (Pontiki et al. 2020;Ndoci 2021a). Presently, references to Albanians are occasionally blatantly xenophobic as in the case of Greek tweets (Pontiki et al. 2020) or a January 2022 comment by Nikos Evangelatos, one of the major news anchors of Greek television, who added Tι πιo σύνηθες απóέναν Aλβανó µε óπλo; "What is more common than an Albanian carrying a gun?" during a news report about a recent arrest of an Albanian man. Stereotypical presentations, however, take often the form of microaggression which makes them harder to identify. ...
Ethnolects have been defined as varieties linked to particular ethnic minorities by the minorities themselves or by other ethnic communities. The present paper investigates this association between ethnic groups and language varieties in the Greek context. I seek to answer whether there is an association made (by Albanians or Greeks) between Albanian migrants in Greece and a particular variety that is not their L1, i.e., Albanian, and if so, whether this is an Albanian ethnolect of Greek. I show experimentally that, in fact, there is a variety of Greek that is linked with listeners’ perceptions of Albanian migrants. However, that criterion is not enough in itself to designate the variety as an ethnolect as the acquisition of this variety by the second or subsequent generations of migrants is not evidenced. Rather, those generations are undergoing language shift from Albanian to Greek. Therefore, the classification of Albanian Greek as an Albanian ethnolect of Greek is not possible despite the association between the variety and the particular minority in Greece. Classification as an L2 Greek variety or a Mock Albanian Greek (MAG) variety is instead argued.
... This annotated corpus was used as a seed for an active learning annotation procedure. Pontiki et al. (2020) created a framework for analyzing verbal aggression and applied it to Greek tweets to investigate xenophobic attitudes expressed through verbal attacks. They utilized a list of keywords to retrieve 6,163,355 tweets for the development of a typology of aggressive messages. ...
We introduce a new corpus, named AIKIA, for Offensive Language Detection (OLD) in Modern Greek (EL). EL is a less-resourced language regarding OLD. AIKIA offers free access to annotated data leveraged from EL Twitter and fiction texts using the lexicon of offensive terms, ERIS, that originates from HurtLex. AIKIA has been annotated for offensive values with the Best Worst Scaling (BWS) method, which is designed to avoid problems of categorical and scalar annotation methods. BWS assigns continuous offensive scores in the form of floating point numbers instead of binary arithmetical or categorical values. AIKIA's performance in OLD was tested by fine-tuning a variety of pre-trained language models in a binary classification task. Experimentation with a number of thresholds showed that the best mapping of the continuous values to binary labels should occur at the range [0.5-0.6] of BWS values and that the pre-trained models on EL data achieved the highest Macro-F1 scores. Greek-Media-BERT outperformed all models with a threshold of 0.6 by obtaining a Macro-F1 score of 0.92.
... They performed event analysis e.g., physical attacks and the involved social actors, using news data from 1995 to 2016 and a study on verbal aggressiveness using Twitter data, locating xenophobic stances as expressed by Greeks in social media for the time period 2013-2016. A follow-up work [154] re-examine verbal aggression as an indicator of xenophobic attitudes in Greek Twitter three years later to trace possible changes. ...
English language is in the spotlight of the Natural Language Processing (NLP) community with other languages, like Greek, lagging behind in terms of offered methods, tools and resources. Due to the increasing interest in NLP, in this paper we try to condense research efforts for the automatic processing of Greek language covering the last three decades. In particular, we list and briefly discuss related works, resources and tools, categorized according to various processing layers and contexts. We are not restricted to the modern form of Greek language but also cover Ancient Greek and various Greek dialects. This survey can be useful for researchers and students interested in NLP tasks, Information Retrieval and Knowledge Management for the Greek language.
... The CLARIN:EL Platform consists of two interconnected subsystems: (a) the Repository, a system with all functionalities related to the provision of LRs, i.e., for depositing and documenting LRs, for curating their metadata through a specially designed metadata editor, for storing, sharing, searching, retrieving, and downloading LRs, and (b) the Workbench, a system providing integrated services that perform core Natural Language Processing (NLP) tasks, such as sentence splitting, tokenization, PoS tagging, lemmatization, parsing, chunking, named entity recognition (Prokopidis & Piperidis, 2020), as well as tasks such as text classification and verbal aggression analysis (Pontiki et al., 2020). Moreover, it offers preprocessing services that perform data format and character encoding conversion. ...
This paper presents the CLARIN:EL infrastructure, which comprises three pillars: the language resources and technologies Platform, the Portal and the Knowledge Centre. It serves as a com-prehensive and interoperable environment that supports language-related research in the fields of language technology, language studies, digital humanities, and political and social sciences. The Platform facilitates deposition, curation and sharing of digital language resources (catering for providers’ needs), and access to and automatic processing of these resources (catering for con-sumers’ needs). The Portal offers informative material about CLARIN:EL and support services to the community, including dissemination, awareness raising and training activities. The Knowledge Centre promotes digital literacy in the scientific domains served, by providing infor-mation on studies, educational and training material and publications. This paper discusses the CLARIN:EL pillars, the technical architecture, its design and implementation principles, the functionalities offered to the users, the support activities provided, usage analytics and future steps.
... Twitter particularly serves as a stimulus for xenophobic narratives and has been used to analyse sentiments on societal issues (Tarisayi & Manik 2020a). Research has examined Twitter discourse on xenophobic attacks (van der Vyver 2019, Pontiki et al. 2020). For instance, van der Vyver (2019) analysed 3,784 tweets during the September 2019 Xenophobic attack, highlighting Twitter's potential to reveal insights, including widespread condemnation. ...
Xenophobia is a pressing issue in South Africa, with frequent instances of violence against immigrants. With the rise of social media, platforms like Twitter reflect public sentiment on this matter. This study examines tweets from 2017 to 2022 about xenophobia in South Africa, using NLP, sentiment analysis, and machine learning to understand public feelings and predict potential xenophobic incidents. The findings aim to help policymakers devise strategies to enhance social cohesion and promote a more inclusive society.
... In similar lines, Capozzi et al. (2019) introduced a platform to detect and monitor hate speech against immigrants, Roma and minorities in Italian social media. On a separate paper, Pontiki et al. (2020) analysed verbal aggression and xenophobic attitudes about immigrants in Greek automatically on Twitter. ...
Human migration is an important societal issue with wide-ranging implications, and timely and accurate insights are increasingly needed for understanding the key factors to ensure the well-being of populations. New data sources, such as usage data from mobile phones and applications, remote sensing and satellite images, social media, event and news databases, and financial databases, enable data scientists to collaborate with migration scholars, to equip them with new quantitative tools, to address certain data gaps, and to supply empirical evidence for building and testing theories while complying with ethical requirements. In this book, we provide an overview of the major data sources and link them to migration and mobility in a way accessible to both migration scholars and data scientists, highlighting the relevant issues from multiple aspects, and offering broad social scientific and technical coverage. We describe many case studies about the use of data science in migration and mobility, as well as related areas, such as humanitarian aid. Most importantly, we give a comprehensive treatment of the legal and ethical concerns, discussing surveillance and dataveillance, implications to power structures, and potential misuses of large scale data processing, which need to be addressed to reap the benefits of data science without harming data subjects, or vulnerable groups such as refugees and asylum seekers.
... Pontiki, Papanikolaou, and Papageorgiou (2018) examine different types of verbal attacks against specific target groups of foreigners as an indicator of xenophobic attitudes on Greek Twitter during the economic crisis, providing a quantitative analysis of the main targets and types of attacks over time and a qualitative analysis of the main stereotypes and prejudices regarding these target groups. Replication of the experiment three years later (Pontiki et al. 2020) suggests a reshaping of online aggression against the same targets in the post-crisis era. ...
El principal objetivo de esta tesis es analizar la relación existente entre la representación de migrantes y refugiados transmitida por los medios informativos y el discurso de odio racista y xenófobo que se propaga de manera masiva en redes sociales, pasando por las actitudes hacia la migración que son reflejadas en las encuestas sociológicas y en las propias plataformas sociales. La hipótesis general es que los medios informativos y la imagen de los migrantes y refugiados que transmiten afectan sobre las ideas, pensamientos, afecciones, actitudes y, finalmente, conductas, que tienen los ciudadanos con respecto a esos colectivos. De esta manera, cuando la imagen predominante difundida sea negativa, las actitudes hacia la migración serán cada vez más negativas, incrementándose el rechazo y, con ello, el odio de tipo racista y xenófobo, el que también influirá sobre esas dimensiones, alimentando así la espiral de odio al reforzar las actitudes y conductas discriminatorias. Con estas premisas, de manera específica, en esta tesis se analizan los marcos connotativos visuales de migrantes y refugiados transmitidos por los principales medios del sur de Europa durante la crisis migratoria, así como el nivel de apoyo a los refugiados expresado por los ciudadanos de esa región en encuestas sociológicas, y terminando por identificar tanto los marcos estudiados como esas actitudes de apoyo o rechazo en los mensajes que se propagan a través de redes sociales, prestando una atención especial al rechazo más explícito, que se expresa en forma de discurso de odio anti-inmigración, por entenderlo como posible detonante de los crímenes de odio. Para ello, se desarrolla una estrategia mixta que incluye desde métodos clásicos como el análisis de contenido manual, hasta técnicas computacionales avanzadas como el topic modeling, el aprendizaje automático supervisado o las poblaciones sintéticas. Los principales resultados evidencian la relación entre los marcos mediáticos de migrantes y refugiados y los mensajes que se propagan en redes sociales en forma de marcos de audiencia, sirviendo los negativos de base argumentativa y motivacional sobre la que se construye el discurso de odio más explícito. En esta línea, a nivel general se confirma que los marcos mediáticos y los marcos de audiencia comparten características similares, que los marcos mediáticos negativos parecen ser, a su vez, sobre los que se construyen los discursos de odio anti-inmigración que se expresan a través de redes sociales, y, además, que las actitudes frente a la migración a nivel social parecen mediar ese camino entre el consumo de ciertos marcos mediáticos
negativos y la conducta final en forma de manifestación verbal del odio en línea, ya que se observaron patrones similares en cada uno de los países en los que se analizaron las diferentes dimensiones. A parte de estos hallazgos empíricos, de manera colateral, con esta tesis se aportan nuevos métodos de investigación aplicados al estudio del discurso de odio online y todos los procesos y dimensiones involucradas, que permiten adaptar la investigación sociológica a los retos y amenazas que presentan los nuevos entornos digitales.
This article describes the bias and discrimination that play a role in the lives of those viewed as different, based on color, ethnicity, gender, language, appearance, age, religion, sexual orientation, or country of origin. Bias and discrimination play a major role in the difficulties faced by unserved and underserved populations such as migrants, refugees, asylum seekers, and others across the globe.
ResearchGate has not been able to resolve any references for this publication.