• Contenu principal
  • Menu
OpenEdition Books
  • Accueil
  • Catalogue de 15900 livres
  • Éditeurs
  • Auteurs
  • Facebook
  • X
  • Partager
    • Facebook

    • X

    • Accueil
    • Catalogue de 15900 livres
    • Éditeurs
    • Auteurs
  • Ressources numériques en sciences humaines et sociales

    • OpenEdition
  • Nos plateformes

    • OpenEdition Books
    • OpenEdition Journals
    • Hypothèses
    • Calenda
  • Bibliothèques

    • OpenEdition Freemium
  • Suivez-nous

  • Lettre d’information
OpenEdition Search

Redirection vers OpenEdition Search.

À quel endroit ?
  • Accademia University Press
  • ›
  • Collana dell'Associazione Italiana di Li...
  • ›
  • EVALITA Evaluation of NLP and Speech Too...
  • ›
  • Track “Affect, Hate, and Stance”
  • ›
  • HaSpeeDe: Hate Speech Detection
  • ›
  • Fontana-Unipi @ HaSpeeDe2: Ensemble of T...
  • Accademia University Press
  • Accademia University Press
    Accademia University Press
    Informations sur la couverture
    Table des matières
    Liens vers le livre
    Informations sur la couverture
    Table des matières
    Formats de lecture

    Plan

    Plan détaillé Texte intégral 1. Introduction 2. Related Work 3. System Architecture 4. Experiments 5. Results and Discussion 6. Conclusions Bibliographie Notes de bas de page Auteurs

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    Ce livre est recensé par

    Précédent Suivant
    Table des matières

    Fontana-Unipi @ HaSpeeDe2: Ensemble of Transformers for the Hate Speech Task at Evalita

    Michele Fontana et Giuseppe Attardi

    p. 136-141

    Résumé

    We describe our approach and experiments to tackle Task A of the second edition of HaSpeeDe, within the Evalita 2020 evaluation campaign. The proposed model consists in an ensemble of classifiers built from three variants of a common neural architecture. Each classifier uses contextual representations from transformers trained on Italian texts, fine tuned on the training set of the challenge. We tested the proposed model on the two official test sets, the in-domain test set containing just tweets and the out-of-domain one including also news headlines. Our submissions ranked 4th on the tweets test set and 17th on the second test set.

    Texte intégral Bibliographie Notes de bas de page Auteurs

    Texte intégral

    1. Introduction

    1The spreading of hateful messages on social media has become a serious issue, therefore techniques of hate speech detection have become quite relevant. The goal of the Hate Speech Detection task (Sanguinetti et al. 2020) at Evalita 2020 (Basile et al. 2020) is to improve the automatic detection of hate messages in Italian tweets. The organizers provided to the participants the dataset HaSpeeDe2, which consists of 6,837 Italian tweets, containing, besides the raw text, also hashtags and emojis. The Task A can be cast into a binary classification task: the model has to predict whether a given message contains hate speech or not.

    2Approaches based on transformer models have become quite popular recently and have proved effective in reaching state-of-the-art scores on major NLP tasks such as those of the GLUE benchmark (Wang et al. 2018). With our experiments we try to assess the effectiveness of transformers trained on Italian documents in a task involving Italian texts from different sources. We experiments with both a transformer model trained specifically on Italian tweets and one trained on generic web documents.

    3We combine several instances of classifiers based on these transformers, in order to address the problem of over-fitting due to the small size of the training set.

    4For this edition of the Evalita HaSpeeDe task, the organizers released two test sets, an in-domain one consisting of tweets and an out-of-domain one containing also news headlines.

    5The ensemble model of our official submission achieved a competitive score of 78.03 Macro-F1 on the in-domain test set but did not perform as well on the second test set.

    6We make available the source code for our experiments as Open Source at https://github.com/mikelefonty/Haspeede2.

    2. Related Work

    7The first edition of HaSpeeDe was held in 2018. The results produced during this contest were the starting point of our research. As described in (Bosco et al. 2018), most of the systems were based on neural networks and used word embeddings, such as FastText (Grave et al. 2018) or word2vec (Polignano and Basile 2018) in the first layer of their architecture. The embeddings layer was usually followed by a Recurrent Network or a Convolutional Neural Network to get an internal representation of the input text. This hidden representation was provided as input to a series of dense layers to obtain the final classification result.

    8Over the last couple of years, the trend in approaches to language analysis has changed considerably, as can be seen by examining the models used in competitions like SemEval 2020 OffensEval 2 (Zampieri et al. 2020). In these new models, to get a better text representation, the embedding layer is often replaced by a Transformer (Vaswani et al. 2017) such as BERT (Devlin et al. 2019), RoBERTa (Liu et al. 2019), or Multilingual BERT (Devlin et al. 2019).

    9We followed this trend but we also focused our attention on the problem raised by the small size of the dataset. As mention, transformer models tend to have a high variance with respect to the input dataset, that often leads to overfitting. The authors therefore suggest to implement an ensemble of classifiers to reduce the variance and consequently improve the generalization capabilities of the trained model.

    10In the following, we describe a similar approach based on the Bagging technique (Breiman 1996), where we apply three different transformer-based classifiers to populate the ensemble and to get the final prediction.

    3. System Architecture

    11During the design phase of our classifier, we looked for a transformer trained directly on a significantly large collection of Italian texts and particularly on Italian tweets, in order to compensate for the small size of the training data. We found two possible models based on BERT: AlBERTo (Polignano et al. 2019)1 and DBMDZ.2 The former is trained on TWITA (Basile, Lai, and Sanguinetti 2018), a 191 GB collection of Italian tweets gathered by the authors, and tested on the SENTIPOLC task during the EVALITA 2016 campaign, where it achieved state-of-the-art accuracy in subjectivity, polarity, and irony detection on Italian tweets. We considered this model suitable for hate speech detection, since its source are Italian tweets and the SENTIPOLC task is a classification task similar to ours. DBMDZ instead is trained on a more general domain, from a 13 GB dataset, which includes a dump of Italian Wikipedia and texts from web pages selected from the Opus Corpora.3 We decided to test both transformer models, assessing their performance through a validation phase on a development set.

    12These transformers were used in the input stage of all our architectures, providing contextual embeddings for sentences that were fine tuned during training. We designed three architecture variants, which were employed as the basic building blocks to construct the ensembles:

    • ALB-SINGLE: It consists of a first layer provided by the AlBERTo transformer, followed by a single neuron with a sigmoid activation function.

    • DB-SINGLE: It follows the same structure of ALB-SINGLE; it just replaces AlBERTo with DBMDZ in the first layer.

    • DB-MLP: Compared to DB-SINGLE, it adds a new dense layer, using a ReLU activation function, between the transformer and the output neuron.

    13The final model is an ensemble consisting of a number of instances of each of the above architectures. For each architecture, e.g. ALB-SINGLE, we construct instances in the following way. After initializing the weights randomly within a given interval and generating the training data by applying the bootstrap technique to the original dataset, we start training the model. When that phase is over, we insert the resulting model in the ensemble. We repeat this process several times with different random weights initialization. Note that, due to the random initialization, no two classifiers in the ensemble are identical to each other. More formally, the model consists of N elements,

    Image 10000000000000CA000000106930FB69C96FF16C.jpg

    14where Image 100000000000008B000000116D33A39A7D544944.jpgrepresent, respectively, the number of instances of ALB-SINGLE, DB-SINGLE and DB-MLP classifiers.

    15In retrospect, it might have been worth while to consider instances of the architecture obtained varying them more thoroughly than just in the initial weights, for example, by changing in the hyper-parameters or number of layers.

    Algorithm 1 Classification Algorithm

    Input: t: the tweet to classify.

    Input: Image 10000000000000870000001344BABF09969978C0.jpg: number of classifiers of each type to be sampled.

    Input: Image 100000000000009200000013C4EF4B696615A56E.jpg: number of classifiers of each type in the ensemble.

    Input: nrun: number of desired iterations.

    Output: cfinal: predicted class

    1:

    preds = []

    2:

    for run = 1 to nrun do

    3:

    Image 10000000000000DA000000133EC120D68DD8C700.jpg

    4:

    Image 10000000000000DC000000139013BB5014BDC000.jpg

    5:

    Image 100000000000010400000013A15D1E33ECBF7729.jpg

    6:

    Image 100000000000012000000011C2612331901333E9.jpg

    7:

    Image 10000000000000D7000000134C2D3B386DD20BDA.jpg

    8:

    for cl in Image 100000000000007F00000011A906183B21C2684B.jpg

    9:

    Image 10000000000000A400000013725B382AA5ED0E8B.jpgclassification

    10:

    end for

    11:

    Image 100000000000010800000040F37A8064E3CC4193.jpg

    12:

    end for

    13:

    Image 100000000000013D00000037B2AE7BEA2E9EF186.jpg

    14:

    return cfinal

    16Our classification algorithm is a slight generalization of the most classical one, which collects results from each member of the ensemble and outputs the class which gets the majority of predictions over all iterations. The process, described by Algorithm 1, performs nrun iterations. During the ith iteration, the algorithm starts sampling randomly from the ensemble a given number of instances for each type of classifier (line 3-5) and initializing to 0 the variable class1, which contains the total number of votes that the hate class receives during the iteration (line 7). It then collects the predictions of the selected models on the tweet t (line 8-10). Image 10000000000000630000001363D1EE9CB0485E6B.jpgrepresents the prediction of classifier cl for the tweet t; in particular cl(t) = 1 if and only if cl classifies t as hateful. The output of iteration i is the most predicted class (line 11). The final result of the algorithm is then the class Image 100000000000006900000014C6E48AB476A88B99.jpg, which obtained the most votes over all the nrun iterations (line 13-14). If Image 100000000000004700000012CCDE397E183AE8C5.jpg, it means that the tweet t has been classified as hateful.

    17A simpler variant of the algorithm would be to just add the counts of each class by all classifiers in all iterations and return the class with the highest count. We plan to compare these two approaches in a future work.

    4. Experiments

    18In this section we describe the experiments we performed to tune the hyper-parameters of our model. We will focus on the search to choose the best values for Image 10000000000000800000000C68492ED4BE54EC10.jpg, that is how many instances to select at each iteration in the classification algorithm.

    19Before starting the experiments, we divided the dataset into two disjoint subsets, a development and an internal test set, in the proportion of 80% and 20%, respectively. The split was done by means of Stratified Sampling, according to the distribution of the target variable hs. We applied the Stratified 3-fold-CV technique to validate our model. Given that we are solving a binary classification problem, we picked the Binary Cross Entropy as our loss. We chose AdamW as our optimizer; we set the first 10% of the total steps as warmup steps. We conducted the experiences on a GPU offered by Google Colab.4 Our models are implemented in PyTorch (Paszke et al. 2019). To extract as much information as possible from input texts, we preprocessed them through hashtag segmentation by means of Tweet Preprocessor.5 We also converted emojis into their Italian description by using the emoji6 and Google Translate7 libraries.

    Table 1: Results of the experiments comparing the baseline architectures. We report the expected value and the standard deviation of the F1 score computed with respect to the 3 validation folds.

    Classifier

    Macro-F1

    Std

    (lr)1-3 ALB-SINGLE

    76.896

    0.7266

    DB-SINGLE

    77.613

    0.3251

    DB-MLP

    78.562

    0.521

    20We analyzed the behaviour of the three baseline architectures we planned to include in the ensemble.

    21We trained each model for a maximum of 4 epochs, using a batch of size 16 and setting the maximum text length to 100. A grid search revealed that the optimal learning rate for DB-MLP is Image 100000000000003800000010EF729A9F413F1339.jpg, and Image 100000000000003800000010233CD7F9691BAA56.jpgfor the remaining models. The optimal number of neurons in the hidden layer of DB-MLP is 50.

    22Table 1 highlights the following aspect: DB-SINGLE achieves better performance than ALB-SINGLE, even though the dataset used to train AlBERTo was composed by a large collection of tweets. The obtained values of the macro-F1 are the baselines of our work.

    23We then describe the results obtained through the ensemble model. To build the classifier, we trained 30 instances of each architecture, keeping the same hyper-parameters obtained from the previous grid search. We thus set:

    Image 10000000000000D00000001073B58C0601C1CC71.jpg

    24We noted that the generalization capability of the ensemble is strictly related to the triple Image 1000000000000087000000131F8350BCA434F26C.jpg, so we performed another grid search, looking for the optimal combination of the three parameters. Table 2 shows the five best configurations found by this search. The optimal values for the triple, (20,25,30), allow the ensemble to achieve an F1-score of 80.0%, with a gain of about 2 points with respect to the score by a single DB-MLP (see Table 1).

    Table 2: Ranking of the 5 best configurations we found, varying the number the number of instances selected from the ensemble. nDB stands for the number of instances of the DB-SINGLE model, and similarly for nMLP and nAL. We report the expected value and the standard deviation of the F1 score computed with respect to the 3 validation folds.

    nDB

    nMLP

    nAL

    Macro-F1

    Std

    (lr)1-5 20

    25

    30

    80.057

    0.534

    15

    20

    25

    80.038

    0.580

    15

    30

    30

    80.036

    0.585

    15

    25

    30

    80.026

    0.563

    15

    30

    15

    80.020

    0.481

    Table 3: Scores by each architecture, both individually and together in the ensemble. We report the average value and the standard deviation of the F1 score computed with respect to the 3 validation folds

    nDB

    nMLP

    nAL

    Macro-F1

    Std

    (lr)1-5 30

    0

    0

    79.074

    0.300

    0

    30

    0

    79.581

    0.3787

    0

    0

    30

    79.482

    0.596

    30

    30

    30

    79.832

    0.525

    Table 4: Results of the final model on the internal test set.

    Accuracy

    Precision

    Recall

    F1

    (lr)1-4 79.313

    78.510

    78.685

    78.592

    25We analyzed the contribution of each architecture individually to the ensemble combination. As shown in Table 3, the best results are obtained with instances of all three architectures. Nevertheless, the results presented in Table 2, show that a more balanced combination achieves better accuracy.

    26We picked the first configuration from Table 2 for our final model and tested it on the internal test set, obtaining the results shown in Table 4.

    5. Results and Discussion

    27The results of our final model applied to the data of the two official test sets of the competition are shown in Table 5. The model performs pretty well on the in-domain dataset, reaching the 4th position in the rankings. However, it did not rank as well in detecting hate speech on the out-of-domain dataset, obtaining an F1-score of just 65.46. The low recall for the hate class highlights that the model fails too often to identify news headlines containing some form of hate speech. In comparison with the official top rankings, listed in Table 6, our model achieved about 12 points below the top score of 77.44\% F1.

    Table 5 : Results of the submitted model on the official blind test stes

    NOT HATE

    HATE

    Precision

    Recall

    F1

    Precision

    Recall

    F1

    Macro-F1

    Position

    Tweets

    81.93

    72.85

    77.12

    74.89

    83.44

    78.94

    78.03

    4

    News

    71.88

    99.37

    83.42

    96.61

    31.49

    47.50

    65.46

    17

    Table 6: Comparison between our final results and the top-5 F1-scores. The values are taken from the official rankings

    Tweets

    News

    Position

    F1 score

    Position

    F1 score

    1

    80.88

    1

    77.44

    2

    78.97

    2

    73.14

    3

    78.93

    3

    72.56

    4

    78.03 (ours)

    4

    71.83

    5

    77.82

    5

    70.2

    6

    77.66

    17

    65.46 (ours)

    28Surprised by this fact, we investigated more deeply, looking for an explanation for such poor result on the out-of-domain dataset.

    29We randomly sampled from the test set some hateful headlines missed by the model, some of which are shown in Table 7.

    Table 7: Examples of hateful headlines, randomly picked from the out-of-domain test set, that are misclassified by our model

    Hateful News Headlines

    anziana rapinata sull’autobus, i due nomadi in fuga si rifugiano al campo di via Candoni

    (elderly woman robbed on the bus, the two fleeing nomads take refuge at the camp on via Candoni)

    Expo: Bordonali, richiedenti asilo in campo base simbolo fallimento governo.

    (Expo: Bordonali, asylum seekers in base camp government failure symbol.)

    Il cardinale M¨uller: ”non possiamo pregare come o con i musulmani”

    ("we cannot pray like nor with Muslims")

    Salvini: ”Il calcio? Rimpiango i tre stranieri in campo”

    (Salvini: "Soccer? I regret the three foreigners on the field")

    30In these headlines, the qualification as hate is implicit and harder to recognize, since it seems due more to the presence of stereotypes (nomads, asylum seekers, Muslims, foreigners), than to the presence of explicit hate expressions.

    31Broadly speaking, we identified some possible reasons for the difference in performance across the two test sets:

    • Linguistic register: Tweets often exhibit a more informal and colloquial language, while headlines employ a more formal lexicon and a more objective tone. This is a crucial difference in identifying hateful messages: while in tweets the feeling of hatred transpires clearly and directly, in headlines this message is conveyed in a more subtle way, often alluding to concepts from political propaganda or common stereotypes. Prior knowledge about the subject and inference might be necessary to decipher the presence of hate. Examining the entire body of the article might have been helpful.

    • Length of text: Tweets are usually longer than news headlines. Thus, the model has fewer elements to exploit to correctly classify a piece of news.

    32These difficulties seem to be shared with other submissions which all got lower scores on the out-of-domain dataset. We expected that pretrained contextual embedding would be more effective in addressing the domain adaptation issue. Further experiments would be needed to improve the resilience of our model.

    6. Conclusions

    33We described an ensemble of neural classifiers, relying on contextual embeddings from transformers, for automated detection of hateful content in Italian texts. We presented the general architecture of our base classification models and how they were combined into an ensemble through a bagging technique. We performed extensive experiments to tune our models and the ensemble on a validation test set. The results achieved by our ensemble model on the in-domain test set confirm its ability in detecting hateful tweets; however the same model performed poorly on the out-of-domain dataset, showing particularly an inability to adapt to handling news headlines. We plan to investigate this issue in future research.

    Julian Risch and Ralf Krestel. 2020. Bagging BERT models for robust aggression identification. In Ritesh Kumar, Atul Kr. Ojha, Bornini Lahiri, Marcos Zampieri, Shervin Malmasi, Vanessa Murdock, and Daniel Kadar, editors, Proceedings of the Second Workshop on Trolling, Aggression and Cyber-bullying, TRAC@LREC 2020, Marseille, France, May 2020, pages 55–61. European Language Re-sources Association (ELRA).

    Bibliographie

    Des DOI sont automatiquement ajoutés aux références bibliographiques par Bilbo, l’outil d’annotation bibliographique d’OpenEdition. Ces références bibliographiques peuvent être téléchargées dans les formats APA, Chicago et MLA.

    Format

    • APA
    • Chicago
    • MLA
    Breiman, L. (1996). Bagging predictors. Machine Learning, 24(2), 123-140. https://doi.org/10.1007/bf00058655
    Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019). Presented at the Proceedings of the 2019 Conference of the North. https://doi.org/10.18653/v1/n19-1423
    Caselli, T., Novielli, N., Patti, V., & Rosso, P. (Eds.). (2018). EVALITA Evaluation of NLP and Speech Tools for Italian. Accademia University Press. https://doi.org/10.4000/books.aaccademia.4421
    Basile, V., Croce, D., Maro, M., & Passaro, L. C. (Eds.). (2020). EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020. Accademia University Press. https://doi.org/10.4000/books.aaccademia.6732
    Wang, A., Singh, A., Michael, J., Hill, F., Levy, O., & Bowman, S. (2018). GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding. Presented at the Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP. https://doi.org/10.18653/v1/w18-5446
    Breiman, Leo. “Bagging Predictors”. Machine Learning 24, nos. 2 (August 1996): 123-40. doi:10.1007/bf00058655.
    Devlin, Jacob, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Proceedings of the 2019 Conference of the North. Association for Computational Linguistics, 2019. doi:10.18653/v1/n19-1423.
    Caselli, Tommaso, Nicole Novielli, Viviana Patti, and Paolo Rosso, eds. EVALITA Evaluation of NLP and Speech Tools for Italian. []. Accademia University Press, 2018. doi:10.4000/books.aaccademia.4421.
    Basile, Valerio, Danilo Croce, Maria Maro, and Lucia C. Passaro, eds. EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020. []. Accademia University Press, 2020. doi:10.4000/books.aaccademia.6732.
    Wang, Alex, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel Bowman. “GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding”. Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP. Association for Computational Linguistics, 2018. doi:10.18653/v1/w18-5446.
    Breiman, Leo. “Bagging Predictors”. Machine Learning, vols. 24, nos. 2, Aug. 1996, pp. 123-40. Crossref, https://doi.org/10.1007/bf00058655.
    Devlin, Jacob, et al. Proceedings of the 2019 Conference of the North, Association for Computational Linguistics, 2019. Crossref, https://doi.org/10.18653/v1/n19-1423.
    Caselli, Tommaso, et al., editors. EVALITA Evaluation of NLP and Speech Tools for Italian. [], Accademia University Press, 2018. Crossref, https://doi.org/10.4000/books.aaccademia.4421.
    Basile, Valerio, et al., editors. EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020. [], Accademia University Press, 2020. Crossref, https://doi.org/10.4000/books.aaccademia.6732.
    Wang, Alex, et al. “GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding”. Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, Association for Computational Linguistics, 2018. Crossref, https://doi.org/10.18653/v1/w18-5446.

    Cette bibliographie a été enrichie de toutes les références bibliographiques automatiquement générées par Bilbo en utilisant Crossref.

    Valerio Basile, Danilo Croce, Maria Di Maro, and Lucia C. Passaro. 2020. “EVALITA 2020: Overview of the 7th Evaluation Campaign of Natural Language Processing and Speech Tools for Italian.” In Proceedings of Seventh Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (Evalita 2020), edited by Valerio Basile, Danilo Croce, Maria Di Maro, and Lucia C. Passaro. Online: CEUR.org.

    Valerio Basile, Mirko Lai, and Manuela Sanguinetti. 2018. “Long-Term Social Media Data Collection at the University of Turin.” In Proceedings of the Fifth Italian Conference on Computational Linguistics (Clic-It 2018), Torino, Italy, December 10-12, 2018, edited by Elena Cabrio, Alessandro Mazzei, and Fabio Tamburini. Vol. 2253. CEUR Workshop Proceedings. CEUR-WS.org. http://ceur-ws.org/Vol-2253/paper48.pdf.

    Cristina Bosco, Felice Dell’Orletta, Fabio Poletto, Manuela Sanguinetti, and Maurizio Tesconi. 2018. “Overview of the EVALITA 2018 Hate Speech Detection Task.” In Proceedings of the Sixth Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (EVALITA 2018) Co-Located with the Fifth Italian Conference on Computational Linguistics (Clic-It 2018), Turin, Italy, December 12-13, 2018, edited by Tommaso Caselli, Nicole Novielli, Viviana Patti, and Paolo Rosso. Vol. 2263. CEUR Workshop Proceedings. CEUR-WS.org. http://ceur-ws.org/Vol-2263/paper010.pdf.

    L. Breiman, 1996. “Bagging Predictors.” Machine Learning 24: 123–40.

    10.1007/BF00058655 :

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. “BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding.” In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, Mn, Usa, June 2-7, 2019, Volume 1 (Long and Short Papers), edited by Jill Burstein, Christy Doran, and Thamar Solorio, 4171–86. Association for Computational Linguistics. https://doi.org/10.18653/v1/n19-1423.

    10.18653/v1/n19-1423 :

    Edouard Grave, Piotr Bojanowski, Prakhar Gupta, Armand Joulin, and Tomas Mikolov. 2018. “Learning Word Vectors for 157 Languages.” In Proceedings of the International Conference on Language Resources and Evaluation (Lrec 2018).

    Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. “RoBERTa: A Robustly Optimized BERT Pretraining Approach.” CoRR abs/1907.11692. http://arxiv.org/abs/1907.11692.

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, et al. 2019. “PyTorch: An Imperative Style, High-Performance Deep Learning Library.” In Advances in Neural Information Processing Systems 32, edited by H. Wallach, H. Larochelle, A. Beygelzimer, F. dAlché-Buc, E. Fox, and R. Garnett, 8024–35. Curran Associates, Inc. http://papers.neurips.cc/paper/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf.

    Marco Polignano, and Pierpaolo Basile. 2018. “HanSEL: Italian Hate Speech Detection Through Ensemble Learning and Deep Neural Networks.” In Proceedings of the Sixth Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (EVALITA 2018) Co-Located with the Fifth Italian Conference on Computational Linguistics (Clic-It 2018), Turin, Italy, December 12-13, 2018, edited by Tommaso Caselli, Nicole Novielli, Viviana Patti, and Paolo Rosso. Vol. 2263. CEUR Workshop Proceedings. CEUR-WS.org. http://ceur-ws.org/Vol-2263/paper038.pdf.

    10.4000/books.aaccademia.4421 :

    Marco Polignano, Pierpaolo Basile, Marco de Gemmis, Giovanni Semeraro, and Valerio Basile. 2019. “AlBERTo: Italian BERT Language Understanding Model for NLP Challenging Tasks Based on Tweets.” In Proceedings of the Sixth Italian Conference on Computational Linguistics, Bari, Italy, November 13-15, 2019, edited by Raffaella Bernardi, Roberto Navigli, and Giovanni Semeraro. Vol. 2481. CEUR Workshop Proceedings. CEUR-WS.org. http://ceur-ws.org/Vol-2481/paper57.pdf.

    Manuela Sanguinetti, Comandini Gloria, Elisa Di Nuovo, Simona Frenda, Marco Stranisci, Cristina Bosco, Tommaso Caselli, Patti Viviana, and Irene Russo. 2020. “HaSpeeDe 2@EVALITA2020: Overview of the EVALITA 2020 Hate Speech Detection Task.” In Proceedings of Seventh Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (EVALITA 2020), edited by Valerio Basile, Danilo Croce, Maria Di Maro, and Lucia C. Passaro. Online: CEUR.org.

    10.4000/books.aaccademia.6732 :

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. “Attention Is All You Need.” In Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, 4-9 December 2017, Long Beach, ca, USA, edited by Isabelle Guyon, Ulrike von Luxburg, Samy Bengio, Hanna M. Wallach, Rob Fergus, S. V. N. Vishwanathan, and Roman Garnett, 5998–6008. http://papers.nips.cc/paper/7181-attention-is-all-you-need.

    Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel Bowman. 2018. “GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding.” In Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, 353–55. Brussels, Belgium: Association for Computational Linguistics. https://doi.org/10.18653/v1/W18-5446.

    10.18653/v1/W18-5446 :

    Marcos Zampieri, Preslav Nakov, Sara Rosenthal, Pepa Atanasova, Georgi Karadzhov, Hamdy Mubarak, Leon Derczynski, Zeses Pitenis, and Çagri Çöltekin. 2020. “SemEval-2020 Task 12: Multilingual Offensive Language Identification in Social Media (Offenseval 2020).” CoRR abs/2006.07235. https://arxiv.org/abs/2006.07235.

    Notes de bas de page

    1 https://github.com/marcopoli/AlBERTo-it

    2 https://huggingface.co/dbmdz/bert-base-italian-uncased

    3 http://opus.nlpl.eu/

    4 https://colab.research.google.com/

    5 https://pypi.org/project/tweet-preprocessor/

    6 https://pypi.org/project/emoji

    7 https://pypi.org/project/googletrans/

    Auteurs

    • Michele Fontana

      Dipartimento di Informatica, Università di Pisa – m.fontana12@studenti.unipi.it

    • Giuseppe Attardi

      Dipartimento di Informatica, Università di Pisa – attardi@di.unipi.it

    Précédent Suivant
    Table des matières

    Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0

    Le texte seul est utilisable sous licence Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

    Voir plus de livres
    Proceedings of the Second Italian Conference on Computational Linguistics CLiC-it 2015

    Proceedings of the Second Italian Conference on Computational Linguistics CLiC-it 2015

    3-4 December 2015, Trento

    Cristina Bosco, Sara Tonelli et Fabio Massimo Zanzotto (dir.)

    2015

    Proceedings of the Third Italian Conference on Computational Linguistics CLiC-it 2016

    Proceedings of the Third Italian Conference on Computational Linguistics CLiC-it 2016

    5-6 December 2016, Napoli

    Anna Corazza, Simonetta Montemagni et Giovanni Semeraro (dir.)

    2016

    EVALITA. Evaluation of NLP and Speech Tools for Italian

    EVALITA. Evaluation of NLP and Speech Tools for Italian

    Proceedings of the Final Workshop 7 December 2016, Naples

    Pierpaolo Basile, Franco Cutugno, Malvina Nissim et al. (dir.)

    2016

    Proceedings of the Fourth Italian Conference on Computational Linguistics CLiC-it 2017

    Proceedings of the Fourth Italian Conference on Computational Linguistics CLiC-it 2017

    11-12 December 2017, Rome

    Roberto Basili, Malvina Nissim et Giorgio Satta (dir.)

    2017

    Proceedings of the Fifth Italian Conference on Computational Linguistics CLiC-it 2018

    Proceedings of the Fifth Italian Conference on Computational Linguistics CLiC-it 2018

    10-12 December 2018, Torino

    Elena Cabrio, Alessandro Mazzei et Fabio Tamburini (dir.)

    2018

    EVALITA Evaluation of NLP and Speech Tools for Italian

    EVALITA Evaluation of NLP and Speech Tools for Italian

    Proceedings of the Final Workshop 12-13 December 2018, Naples

    Tommaso Caselli, Nicole Novielli, Viviana Patti et al. (dir.)

    2018

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    Proceedings of the Seventh Evaluation Campaign of Natural Language Processing and Speech Tools for Italian Final Workshop

    Valerio Basile, Danilo Croce, Maria Maro et al. (dir.)

    2020

    Proceedings of the Seventh Italian Conference on Computational Linguistics CLiC-it 2020

    Proceedings of the Seventh Italian Conference on Computational Linguistics CLiC-it 2020

    Bologna, Italy, March 1-3, 2021

    Felice Dell'Orletta, Johanna Monti et Fabio Tamburini (dir.)

    2020

    Proceedings of the Eighth Italian Conference on Computational Linguistics CliC-it 2021

    Proceedings of the Eighth Italian Conference on Computational Linguistics CliC-it 2021

    Milan, Italy, 26-28 January, 2022

    Elisabetta Fersini, Marco Passarotti et Viviana Patti (dir.)

    2022

    Voir plus de livres
    1 / 9
    Proceedings of the Second Italian Conference on Computational Linguistics CLiC-it 2015

    Proceedings of the Second Italian Conference on Computational Linguistics CLiC-it 2015

    3-4 December 2015, Trento

    Cristina Bosco, Sara Tonelli et Fabio Massimo Zanzotto (dir.)

    2015

    Proceedings of the Third Italian Conference on Computational Linguistics CLiC-it 2016

    Proceedings of the Third Italian Conference on Computational Linguistics CLiC-it 2016

    5-6 December 2016, Napoli

    Anna Corazza, Simonetta Montemagni et Giovanni Semeraro (dir.)

    2016

    EVALITA. Evaluation of NLP and Speech Tools for Italian

    EVALITA. Evaluation of NLP and Speech Tools for Italian

    Proceedings of the Final Workshop 7 December 2016, Naples

    Pierpaolo Basile, Franco Cutugno, Malvina Nissim et al. (dir.)

    2016

    Proceedings of the Fourth Italian Conference on Computational Linguistics CLiC-it 2017

    Proceedings of the Fourth Italian Conference on Computational Linguistics CLiC-it 2017

    11-12 December 2017, Rome

    Roberto Basili, Malvina Nissim et Giorgio Satta (dir.)

    2017

    Proceedings of the Fifth Italian Conference on Computational Linguistics CLiC-it 2018

    Proceedings of the Fifth Italian Conference on Computational Linguistics CLiC-it 2018

    10-12 December 2018, Torino

    Elena Cabrio, Alessandro Mazzei et Fabio Tamburini (dir.)

    2018

    EVALITA Evaluation of NLP and Speech Tools for Italian

    EVALITA Evaluation of NLP and Speech Tools for Italian

    Proceedings of the Final Workshop 12-13 December 2018, Naples

    Tommaso Caselli, Nicole Novielli, Viviana Patti et al. (dir.)

    2018

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    Proceedings of the Seventh Evaluation Campaign of Natural Language Processing and Speech Tools for Italian Final Workshop

    Valerio Basile, Danilo Croce, Maria Maro et al. (dir.)

    2020

    Proceedings of the Seventh Italian Conference on Computational Linguistics CLiC-it 2020

    Proceedings of the Seventh Italian Conference on Computational Linguistics CLiC-it 2020

    Bologna, Italy, March 1-3, 2021

    Felice Dell'Orletta, Johanna Monti et Fabio Tamburini (dir.)

    2020

    Proceedings of the Eighth Italian Conference on Computational Linguistics CliC-it 2021

    Proceedings of the Eighth Italian Conference on Computational Linguistics CliC-it 2021

    Milan, Italy, 26-28 January, 2022

    Elisabetta Fersini, Marco Passarotti et Viviana Patti (dir.)

    2022

    Voir plus de chapitres

    Detecting the scope of negations in clinical notes

    Giuseppe Attardi, Vittoria Cozza et Daniele Sartiano

    Deep Learning for Social Sensing from Tweets

    Giuseppe Attardi, Laura Gorrieri, Alessio Miaschi et al.

    Evolution of Italian Treebank and Dependency Parsing towards Universal Dependencies

    Giuseppe Attardi, Simone Saletti et Maria Simi

    Using Embeddings for Both Entity Recognition and Linking in Tweets

    Giuseppe Attardi, Daniele Sartiano, Maria Simi et al.

    Character Embeddings PoS Tagger vs HMM Tagger for Tweets

    Giuseppe Attardi et Maria Simi

    Convolutional Neural Networks for Sentiment Analysis on Italian Tweets

    Giuseppe Attardi, Daniele Sartiano, Chiara Alzetta et al.

    Voir plus de chapitres
    1 / 6

    Detecting the scope of negations in clinical notes

    Giuseppe Attardi, Vittoria Cozza et Daniele Sartiano

    Deep Learning for Social Sensing from Tweets

    Giuseppe Attardi, Laura Gorrieri, Alessio Miaschi et al.

    Evolution of Italian Treebank and Dependency Parsing towards Universal Dependencies

    Giuseppe Attardi, Simone Saletti et Maria Simi

    Using Embeddings for Both Entity Recognition and Linking in Tweets

    Giuseppe Attardi, Daniele Sartiano, Maria Simi et al.

    Character Embeddings PoS Tagger vs HMM Tagger for Tweets

    Giuseppe Attardi et Maria Simi

    Convolutional Neural Networks for Sentiment Analysis on Italian Tweets

    Giuseppe Attardi, Daniele Sartiano, Chiara Alzetta et al.

    Accès ouvert

    Accès ouvert

    ePub

    PDF

    PDF du chapitre

    1 https://github.com/marcopoli/AlBERTo-it

    2 https://huggingface.co/dbmdz/bert-base-italian-uncased

    3 http://opus.nlpl.eu/

    4 https://colab.research.google.com/

    5 https://pypi.org/project/tweet-preprocessor/

    6 https://pypi.org/project/emoji

    7 https://pypi.org/project/googletrans/

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    X Facebook Email

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    Vous pouvez vous connecter à votre bibliothèque à l’adresse suivante : https://freemium.openedition.org/oebooks

    Suggérer l’acquisition à votre bibliothèque

    Si vous avez des questions, vous pouvez nous écrire à access[at]openedition.org

    EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020

    Vérifiez si votre bibliothèque a déjà acquis ce livre : authentifiez-vous à OpenEdition Freemium for Books.

    Vous pouvez suggérer à votre bibliothèque d’acquérir un ou plusieurs livres publiés sur OpenEdition Books. N’hésitez pas à lui indiquer nos coordonnées : access[at]openedition.org

    Vous pouvez également nous indiquer, à l’aide du formulaire suivant, les coordonnées de votre bibliothèque afin que nous la contactions pour lui suggérer l’achat de ce livre. Les champs suivis de (*) sont obligatoires.

    Veuillez, s’il vous plaît, remplir tous les champs.

    La syntaxe de l’email est incorrecte.

    Référence numérique du chapitre

    Format

    Fontana, M., & Attardi, G. (2020). Fontana-Unipi @ HaSpeeDe2: Ensemble of Transformers for the Hate Speech Task at Evalita. In V. Basile, D. Croce, M. Maro, & L. C. Passaro (éds.), EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020. Torino: Accademia University Press. https://doi.org/10.4000/books.aaccademia.6979
    Fontana, Michele, et Giuseppe Attardi. « Fontana-Unipi @ HaSpeeDe2: Ensemble of Transformers for the Hate Speech Task at Evalita ». In EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020, édité par Valerio Basile, Danilo Croce, Maria Maro, et Lucia C. Passaro. Torino: Accademia University Press, 2020. doi:10.4000/books.aaccademia.6979.
    Fontana, Michele, et Giuseppe Attardi. « Fontana-Unipi @ HaSpeeDe2: Ensemble of Transformers for the Hate Speech Task at Evalita ». EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020, édité par Valerio Basile et al., Accademia University Press, 2020, https://doi.org/10.4000/books.aaccademia.6979.

    Référence numérique du livre

    Format

    Basile, V., Croce, D., Maro, M., & Passaro, L. C. (éds.). (2020). EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020. Torino: Accademia University Press. https://doi.org/10.4000/books.aaccademia.6732
    Basile, Valerio, Danilo Croce, Maria Maro, et Lucia C. Passaro, éd. EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020. Torino: Accademia University Press, 2020. doi:10.4000/books.aaccademia.6732.
    Basile, Valerio, et al., éditeurs. EVALITA Evaluation of NLP and Speech Tools for Italian - December 17th, 2020. Accademia University Press, 2020, https://doi.org/10.4000/books.aaccademia.6732.
    Compatible avec Zotero Zotero

    1 / 3

    Accademia University Press

    Accademia University Press

    • Plan du site
    • Se connecter

    Suivez-nous

    • Facebook
    • Flux RSS

    URL : http://www.aaccademia.it/

    Email : info@aaccademia.it

    Adresse :

    Accademia University Press

    Via Carlo Alberto 55

    I‐10123

    Torino

    Italia

    OpenEdition
    • Candidater à OpenEdition Books
    • Connaître le programme OpenEdition Freemium
    • Commander des livres
    • S’abonner à la lettre d’OpenEdition
    • CGU d’OpenEdition Books
    • Accessibilité : partiellement conforme
    • Données personnelles
    • Gestion des cookies
    • Système de signalement