The paper introduces a novel corpus collected in a set of experiments in Italian schools, annotated for the presence of stereotypes, and related categories. It consists of comments written by teenage students in reaction to fabricated fake news, designed to elicit prejudiced responses, by featuring racial stereotypes. We make use of an annotation scheme which takes into account the implicit or explicit nature of different instances of stereotypes, alongside their forms of discredit. We also annotate the stance of the commenter towards the news article, using a schema inspired by rumor and fake news stance detection tasks. Through this rarely studied setting, we provide a preliminary exploration of the production of stereotypes in a more controlled context. Alongside this novel dataset, we provide both quantitative and qualitative analyses of these reactions, to validate the categories used in their annotation. Through this work, we hope to increase the diversity of available data in the study of the propagation and the dynamics of negative stereotypes.

The paper describes a dataset composed of two sub-corpora from two different sources in Italian. The QUEEREOTYPES corpus includes social media texts regarding LGBTQIA+ individuals, behaviors, ideology and events. The texts were collected from Facebook and Twitter in 2018 and were annotated for the presence of stereotypes, and orthogonal dimensions (such as hate speech, aggressiveness, offensiveness, and irony in one sub-corpus, and stance in the other). The resource was developed by Natural Language Processing researchers together with activists from an Italian LGBTQIA+ not-for-profit organization. The creation of the dataset allows the NLP community to study stereotypes against marginalized groups, individuals and, ultimately, to develop proper tools and measures to reduce the online spread of such stereotypes. A test for the robustness of the language resource has been performed by means of 5-fold cross-validation experiments. Finally, text classification experiments have been carried out with a fine-tuned version of AlBERTo (a BERT-based model pre-trained on Italian tweets) and mBERT, obtaining good results on the task of stereotype detection, suggesting that stereotypes towards different targets might share common traits.

Research in the field of NLP has recently focused on the variability that people show in selecting labels when performing an annotation task. Exploiting disagreements in annotations has been shown to offer advantages for accurate modelling and fair evaluation. In this paper, we propose a strongly perspectivist model for supervised classification of natural language utterances. Our approach combines the predictions of several perspective-aware models using key information of their individual confidence to capture the subjectivity encoded in the annotation of linguistic phenomena. We validate our method through experiments on two case studies, irony and hate speech detection, in in-domain and cross-domain settings. The results show that confidence-based ensembling of perspective-aware models seems beneficial for classification performance in all scenarios. In addition, we demonstrate the effectiveness of our method with automatically extracted perspectives from annotations when the annotators’ metadata are not available.

In this paper, we focus on the topics of misinformation and racial hoaxes from a perspective derived from both social psychology and computational linguistics. In particular, we consider the specific case of anti-immigrant feeling as a first case study for addressing racial stereotypes. We describe the first corpus-based study for multilingual racial stereotype identification in social media conversational threads. Our contributions are: (i) a multilingual corpus of racial hoaxes, (ii) a set of common guidelines for the annotation of racial stereotypes in social media texts, and a multi-layered, fine-grained scheme, psychologically grounded on the work by Fiske, including not only stereotype presence, but also contextuality, implicitness, and forms of discredit, (iii) a multilingual dataset in Italian, Spanish, and French annotated following the aforementioned guidelines, and cross-lingual comparative analyses taking into account racial hoaxes and stereotypes in online discussions. The analysis and results show the usefulness of our methodology and resources, shedding light on how racial hoaxes are spread, and enable the identification of negative stereotypes that reinforce them.

We present EPIC (English Perspectivist Irony Corpus), the first annotated corpus for irony analysis based on the principles of data perspectivism. The corpus contains short conversations from social media in five regional varieties of English, and it is annotated by contributors from five countries corresponding to those varieties. We analyse the resource along the perspectives induced by the diversity of the annotators, in terms of origin, age, and gender, and the relationship between these dimensions, irony, and the topics of conversation. We validate EPIC by creating perspective-aware models that encode the perspectives of annotators grouped according to their demographic characteristics. Firstly, the performance of perspectivist models confirms that different annotators induce very different models. Secondly, in the classification of ironic and non-ironic texts, perspectivist models prove to be generally more confident than the non-perspectivist ones. Furthermore, comparing the performance on a perspective-based test set with those achieved on a gold standard test set, we can observe how perspectivist models tend to detect more precisely the positive class, showing their ability to capture the different perceptions of irony. Thanks to these models, we are moreover able to show interesting insights about the variation in the perception of irony by the different groups of annotators, such as among different generations and nationalities.

This paper introduces the Unified Interactive Natural Understanding of the Italian Language (UINAUIL), a benchmark of six tasks for Italian Natural Language Understanding. We present a description of the tasks and software library that collects the data from the European Language Grid, harmonizes the data format, and exposes functionalities to facilitates data manipulation and the evaluation of custom models. We also present the results of tests conducted with available Italian and multilingual language models on UINAUIL, providing an updated picture of the current state of the art in Italian NLU.

Inside the NLP community there is a considerable amount of language resources created, annotated and released every day with the aim of studying specific linguistic phenomena. Despite a variety of attempts in order to organize such resources has been carried on, a lack of systematic methods and of possible interoperability between resources are still present. Furthermore, when storing linguistic information, still nowadays, the most common practice is the concept of “gold standard”, which is in contrast with recent trends in NLP that aim at stressing the importance of different subjectivities and points of view when training machine learning and deep learning methods. In this paper we present O-Dang!: The Ontology of Dangerous Speech Messages, a systematic and interoperable Knowledge Graph (KG) for the collection of linguistic annotated data. O-Dang! is designed to gather and organize Italian datasets into a structured KG, according to the principles shared within the Linguistic Linked Open Data community. The ontology has also been designed to account a perspectivist approach, since it provides a model for encoding both gold standard and single-annotator labels in the KG. The paper is structured as follows. In Section 1 the motivations of our work are outlined. Section 2 describes the O-Dang! Ontology, that provides a common semantic model for the integration of datasets in the KG. The Ontology Population stage with information about corpora, users, and annotations is presented in Section 3. Finally, in Section 4 an analysis of offensiveness across corpora is provided as a first case study for the resource.

The European Language Grid enables researchers and practitioners to easily distribute and use NLP resources and models, such as corpora and classifiers. We describe in this paper how, during the course of our EVALITA4ELG project, we have integrated datasets and systems for the Italian language. We show how easy it is to use the integrated systems, and demonstrate in case studies how seamless the application of the platform is, providing Italian NLP for everyone.

In this paper we present a set of multilingual experiments tackling the task of Stance Detection in five different languages: English, Spanish, Catalan, French and Italian. Furthermore, we study the phenomenon of stance with respect to six different targets – one per language, and two different for Italian – employing a variety of machine learning algorithms that primarily exploit morphological and syntactic knowledge as features, represented throughout the format of Universal Dependencies. Results seem to suggest that the methodology employed is not beneficial per se, but might be useful to exploit the same features with a different methodology.

The recognition of irony is a challenging task in the domain of Sentiment Analysis, and the availability of annotated corpora may be crucial for its automatic processing. In this paper we describe a fine-grained annotation scheme centered on irony, in which we highlight the tokens that are responsible for its activation, (irony activators) and their morpho-syntactic features. As our case study we therefore introduce a recently released Universal Dependencies treebank for Italian which includes ironic tweets: TWITTIRÒ-UD. For the purposes of this study, we enriched the existing annotation in the treebank, with a further level that includes irony activators. A description and discussion of the annotation scheme is provided with a definition of irony activators and the guidelines for their annotation. This qualitative study on the different layers of annotation applied on the same dataset can shed some light on the process of human annotation, and irony annotation in particular, and on the usefulness of this representation for developing computational models of irony to be used for training purposes.

The paper presents a discussion on the main linguistic phenomena of user-generated texts found in web and social media, and proposes a set of annotation guidelines for their treatment within the Universal Dependencies (UD) framework. Given on the one hand the increasing number of treebanks featuring user-generated content, and its somewhat inconsistent treatment in these resources on the other, the aim of this paper is twofold: (1) to provide a short, though comprehensive, overview of such treebanks - based on available literature - along with their main features and a comparative analysis of their annotation criteria, and (2) to propose a set of tentative UD-based annotation guidelines, to promote consistent treatment of the particular phenomena found in these types of texts. The main goal of this paper is to provide a common framework for those teams interested in developing similar resources in UD, thus enabling cross-linguistic consistency, which is a principle that has always been in the spirit of UD.

This paper presents an in-depth investigation of the effectiveness of dependency-based syntactic features on the irony detection task in a multilingual perspective (English, Spanish, French and Italian). It focuses on the contribution from syntactic knowledge, exploiting linguistic resources where syntax is annotated according to the Universal Dependencies scheme. Three distinct experimental settings are provided. In the first, a variety of syntactic dependency-based features combined with classical machine learning classifiers are explored. In the second scenario, two well-known types of word embeddings are trained on parsed data and tested against gold standard datasets. In the third setting, dependency-based syntactic features are combined into the Multilingual BERT architecture. The results suggest that fine-grained dependency-based syntactic information is informative for the detection of irony.

The paper describes the organization of the SemEval 2019 Task 5 about the detection of hate speech against immigrants and women in Spanish and English messages extracted from Twitter. The task is organized in two related classification subtasks: a main binary subtask for detecting the presence of hate speech, and a finer-grained one devoted to identifying further features in hateful contents such as the aggressive attitude and the target harassed, to distinguish if the incitement is against an individual rather than a group. HatEval has been one of the most popular tasks in SemEval-2019 with a total of 108 submitted runs for Subtask A and 70 runs for Subtask B, from a total of 74 different teams. Data provided for the task are described by showing how they have been collected and annotated. Moreover, the paper provides an analysis and discussion about the participant systems and the results they achieved in both subtasks.

In the present paper we describe the UPV-28-UNITO system’s submission to the RumorEval 2019 shared task. The approach we applied for addressing both the subtasks of the contest exploits both classical machine learning algorithms and word embeddings, and it is based on diverse groups of features: stylistic, lexical, emotional, sentiment, meta-structural and Twitter-based. A novel set of features that take advantage of the syntactic information in texts is moreover introduced in the paper.

This paper provides a linguistic and pragmatic analysis of the phenomenon of irony in order to represent how Twitter’s users exploit irony devices within their communication strategies for generating textual contents. We aim to measure the impact of a wide-range of pragmatic phenomena in the interpretation of irony, and to investigate how these phenomena interact with contexts local to the tweet. Informed by linguistic theories, we propose for the first time a multi-layered annotation schema for irony and its application to a corpus of French, English and Italian tweets. We detail each layer, explore their interactions, and discuss our results according to a qualitative and quantitative perspective.

The paper introduces a new annotated French data set for Sentiment Analysis, which is a currently missing resource. It focuses on the collection from Twitter of data related to the socio-political debate about the reform of the bill for wedding in France. The design of the annotation scheme is described, which extends a polarity label set by making available tags for marking target semantic areas and figurative language devices. The annotation process is presented and the disagreement discussed, in particular, in the perspective of figurative language use and in that of the semantic oriented annotation, which are open challenges for NLP systems.

In this paper we present the TWitterBuonaScuola corpus (TW-BS), a novel Italian linguistic resource for Sentiment Analysis, developed with the main aim of analyzing the online debate on the controversial Italian political reform “Buona Scuola” (Good school), aimed at reorganizing the national educational and training systems. We describe the methodologies applied in the collection and annotation of data. The collection has been driven by the detection of the hashtags mainly used by the participants to the debate, while the annotation has been focused on sentiment polarity and irony, but also extended to mark the aspects of the reform that were mainly discussed in the debate. An in-depth study of the disagreement among annotators is included. We describe the collection and annotation stages, and the in-depth analysis of disagreement made with Crowdflower, a crowdsourcing annotation platform.

This paper aims to introduce the issues related to the syntactic alignment of a dependency-based multilingual parallel treebank, ParTUT. Our approach to the task starts from a lexical mapping and then attempts to expand it using dependency relations. In developing the system, however, we realized that the only dependency relations between the individual nodes were not sufficient to overcome some translation divergences, or shifts, especially in the absence of a direct lexical mapping and a different syntactic realization. For this purpose, we explored the use of a novel syntactic notion introduced in dependency theoretical framework, i.e. that of catena (Latin for “chain”), which is intended as a group of words that are continuous with respect to dominance. In relation to the task of aligning parallel dependency structures, catenae can be used to explain and identify those cases of one-to-many or many-to-many correspondences, typical of several translation shifts, that cannot be detected by means of direct word-based mappings or bare syntactic relations. The paper presented here describes the overall structure of the alignment system as it has been currently designed, how catenae are extracted from the parallel resource, and their potential relevance to the completion of tree alignment in ParTUT sentences.

Stanford Dependencies (SD) represent nowadays a de facto standard as far as dependency annotation is concerned. The goal of this paper is to explore pros and cons of different strategies for generating SD annotated Italian texts to enrich the existing Italian Stanford Dependency Treebank (ISDT). This is done by comparing the performance of a statistical parser (DeSR) trained on a simpler resource (the augmented version of the Merged Italian Dependency Treebank or MIDT+) and whose output was automatically converted to SD, with the results of the parser directly trained on ISDT. Experiments carried out to test reliability and effectiveness of the two strategies show that the performance of a parser trained on the reduced dependencies repertoire, whose output can be easily converted to SD, is slightly higher than the performance of a parser directly trained on ISDT. A non-negligible advantage of the first strategy for generating SD annotated texts is that semi-automatic extensions of the training resource are more easily and consistently carried out with respect to a reduced dependency tag set. Preliminary experiments carried out for generating the collapsed and propagated SD representation are also reported.

The paper introduces an ongoing project for the development of a parallel treebank for Italian, English and French, i.e. Parallel--TUT, or simply ParTUT. For the development of this resource, both the dependency and constituency-based formats of the Italian Turin University Treebank (TUT) have been applied to a preliminary dataset, which includes the whole text of the Universal Declaration of Human Rights, and sentences from the JRC-Acquis Multilingual Parallel Corpus and the Creative Commons licence. The focus of the project is mainly on the quality of the annotation and the investigation of some issues related to the alignment of data that can be allowed by the TUT formats, also taking into account the availability of conversion tools for display data in standard ways, such as Tiger--XML and CoNLL formats. It is, in fact, our belief that increasing the portability of our treebank could give us the opportunity to access resources and tools provided by other research groups, especially at this stage of the project, where no particular tool -- compatible with the TUT format -- is available in order to tackle the alignment problems.

The aim of this paper is to contribute to the debate on the issues raised by Morphologically Rich Languages, and more precisely to investigate, in a cross-paradigm perspective, the influence of the constituent order on the data-driven parsing of one of such languages(i.e. Italian). It shows therefore new evidence from experiments on Italian, a language characterized by a rich verbal inflection, which leads to a widespread diffusion of the pro―drop phenomenon and to a relatively free word order. The experiments are performed by using state-of-the-art data-driven parsers (i.e. MaltParser and Berkeley parser) and are based on an Italian treebank available in formats that vary according to two dimensions, i.e. the paradigm of representation (dependency vs. constituency) and the level of detail of linguistic information.

As the interest of the NLP community grows to develop several treebanks also for languages other than English, we observe efforts towards evaluating the impact of different annotation strategies used to represent particular languages or with reference to particular tasks. This paper contributes to the debate on the influence of resources used for the training and development on the performance of parsing systems. It presents a comparative analysis of the results achieved by three different dependency parsers developed and tested with respect to two treebanks for the Italian language, namely TUT and ISST--TANL, which differ significantly at the level of both corpus composition and adopted dependency representations.

Subcategorization is a kind of knowledge which can be considered as crucial in several NLP tasks, such as Information Extraction or parsing, but the collection of very large resources including subcategorization representation is difficult and time-consuming. Various experiences show that the automatic extraction can be a practical and reliable solution for acquiring such a kind of knowledge. The aim of this paper is to investigate the relationships between subcategorization frame extraction and the nature of data from which the frames have to be extracted, e.g. how much the task can be influenced by the richness/poorness of the annotation. Therefore, we present some experiments that apply statistical subcategorization extraction methods, known in literature, on an Italian treebank that exploits a rich set of dependency relations that can be annotated at different degrees of specificity. Benefiting from the availability of relation sets that implement different granularity in the representation of relations, we evaluate our results with reference to previous works in a cross-linguistic perspective.

The EVALITA 2007 Parsing Task has been the first contest among parsing systems for Italian. It is the first attempt to compare the approaches and the results of the existing parsing systems specific for this language using a common treebank annotated using both a dependency and a constituency-based format. The development data set for this parsing competition was taken from the Turin University Treebank, which is annotated both in dependency and constituency format. The evaluation metrics were those standardly applied in CoNLL and PARSEVAL. The results of the parsing results are very promising and higher than the state-of-the-art for dependency parsing of Italian. An analysis of such results is provided, which takes into account other experiences in treebank-driven parsing for Italian and for other Romance languages (in particular, the CoNLL X & 2007 shared tasks for dependency parsing). It focuses on the characteristics of data sets, i.e. type of annotation and size, parsing paradigms and approaches applied also to languages other than Italian.

EVALITA 2007, the first edition of the initiative devoted to the evaluation of Natural Language Processing tools for Italian, provided a shared framework where participants’ systems had the possibility to be evaluated on five different tasks, namely Part of Speech Tagging (organised by the University of Bologna), Parsing (organised by the University of Torino), Word Sense Disambiguation (organised by CNR-ILC, Pisa), Temporal Expression Recognition and Normalization (organised by CELCT, Trento), and Named Entity Recognition (organised by FBK, Trento). We believe that the diffusion of shared tasks and shared evaluation practices is a crucial step towards the development of resources and tools for Natural Language Processing. Experiences of this kind, in fact, are a valuable contribution to the validation of existing models and data, allowing for consistent comparisons among approaches and among representation schemes. The good response obtained by EVALITA, both in the number of participants and in the quality of results, showed that pursuing such goals is feasible not only for English, but also for other languages.

The paper investigates the issue of portability of methods and results over treebanks in different languages and annotation formats. In particular, it addresses the problem of converting an Italian treebank, the Turin University Treebank (TUT), developed in dependency format, into the Penn Treebank format, in order to possibly exploit the tools and methods already developed and compare the adequacy of information encoding in the two formats. We describe the procedures for converting the two annotation formats and we present an experiment that evaluates some linguistic knowledge extracted from the two formats, namely sub-categorization frames.