English. In this article, we describe two classification models (a Convolutional Neural Network and a Logistic Regression classifier), arranged according to three different strategies, submitted to subtask A of Automatic Misogyny Identification at EVALITA 2020. Results were very encouraging for detecting misogyny, even though aggressiveness was less accurate. Our second strategy, consisting of a Convolutional Neural Network and logistic regression to identify misogyny and aggressiveness, respectively, won the sixth place in the competition.
Italiano. In questo articolo, descriviamo due modelli di classificazione (i.e., Convolutional Neural Network e Regressione Logistica), organizzati secondo tre diverse strategie, per il subtask A dello shared task Automatic Misogyny Identification a EVALITA 2020. I risultati sono stati molto incoraggianti nel rilevamento della misoginia, anche se l’aggressivita` viene riconosciuta con una precisione piu` basse. La nostra seconda strategia (Convolutional Neural Network per misoginia e Regressione Logistica per aggressivita`) ci ha permesso di ottenere il sesto posto nella competizione.
Hate speech is a problem that has been gaining space both in the media and in academic research. Political organizations have been working to combat this type of discourse. As is the case with the
Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). code of conduct1 created by the European Union Commission, and signed by some of the main social networks, such as Facebook, YouTube, Twitter, which aims to monitor and remove this type of content within 24 hours of its disclosure.
The subject has even become a marketing problem, to the extent that recently several companies stopped advertising on Facebook2, only to put some pressure at the network to have it remove this type of publication from the posts within it. Advertisers point, in this case, is that they do not want their brand to be linked to this type of speech.
Defined as “language which attacks or demeans
a group based on race, ethnic origin, religion,
gender, age, disability, or sexual orientation/gender
identity“
When this hate speech is targeted specifically
at women, it is called misogyny
Some academic work and several competitions
have proposed some tasks to promote studies and
advances in the area. Much of this work and
data sets focus on English
It is extremely important, therefore, to encourage the development of this kind of study 1https://ec.europa.eu/info/policies/justice-andfundamental-rights/combatting-discrimination/racismand-xenophobia/eu-code-conduct-countering-illegal-hatespeech-online en
2https://www.nytimes.com/2020/08/01/
business/media/facebook-boycott.html
in different languages and competitions, such as
IberEval
In this work, we help address this problem by testing two classification models as part of EVALITA 2020’s subtask A on Automatic Misogyny Identification (AMI). Tested models were a Convolutional Neural Network (CNN) and a Logistic Regression (LR) classifier. Three different strategies were designed and tested, with one of them scoring 6th in the competition.
The rest of this article is organized as follows. Section 2 presents some related work in the identification of misogyny or hate speech. Section 3, in turn, gives an overview of EVALITA’s AMI. Next, in section 4, we describe our experimental set-up, giving details of the implemented methods and tested strategies. Finally, in Section 5 we discuss our results, whereas in Section 6 we present our final remarks on this task. 2
IberEval
SVMs were also proposed to identify racism
in Twitter messages in English, achieving an F1
score of 0.76
During the Automatic Misogyny Identification
shared task at EVALITA 2018, it was proposed a
subtask A, which consisted of identifying
misogyny
The second edition of misogyny identification
at EVALITA 2020 consists of two subtasks: A
and B. The purpose of subtask A is to
identify the presence or absence of misogyny and
aggressiveness in tweets
The ”No Place For Hate Speech” team
participated only in subtask A, and all discussions that
will be followed are related to this subtask. Within
EVALITA 2020, the subtask consisted of
identifying the presence or absence of misogynous speech
and aggressiveness in tweets in Italian
The training dataset consisted of 5,000 tweets.
The class that determines the presence or absence
of misogyny is nearly balanced. However,
aggressiveness is not balanced at all, with approximately
35% of tweets containing aggressiveness. Table 1
shows the distribution of each class in the training
set.
In subtask A, we tested two different
classifiers within different configurations. These were
a Convolutional Neural Network (CNN), using
BERT
The LR classifier used a 4-gram language
model, with tf-idf
Since the subtask A at EVALITA allows each team to submit up to three classifiers, we decided to approach the problem according to three different strategies, involving different combinations of these classifiers, along with different subsets of data on which they should be trained.
3https://www.tensorflow.org/ 4https://scikit-learn.org/stable/ In all cases, the training set was divided in a 90% subset, used for training purposes, with the remaining 10% used for out-of-sample testing. All classifiers used this same proportion both to identify misogyny and aggressiveness. Tweets were used in their raw form and no preprocessing was used.
All CNNs used in the experiments had the same configuration, being trained for 15 epochs. They also have three convolution layers, relu activation functions, and dropout rate of 0.10, with adam optimisation. Finally, cross-entropy was used as their loss function. In what follows, we will describe, with more details, each of the strategies followed during our tests.
The first strategy consisted of training two CNNs, one for each specific sub-problem separately, i.e. one for misogyny and another for aggressiveness classification. In both cases, the entire data set was used for training.
At the testing stage, the CNNs were arranged as a pipeline, in which the first CNN was responsible for identifying whether a tweet had some misogynous content, whereas the second CNN was responsible for identifying the presence or absence of aggressiveness only in those tweets marked as misogynous by the first CNN.
Similar to Strategy 1, the second strategy also consisted of training a CNN to detect misogynous content in tweets. This time, however, the classification of aggressiveness was left to a Linear Regression classifier. As in the first strategy, both models were trained in the entire data set.
During testing, once again models were arranged in a pipeline, with the CNN coming first, to detect misogyny in tweets. In the sequence, all tweets classified as misogynous by the CNN were then fed to the LR classifier, so it could determine the presence or absence of aggressiveness.
Our third strategy is similar to Strategy 1, in that it also consists of two CNNs trained separately over the data set. The only difference, however, lies during the training stage. In this case, whereas the first CNN (i.e. the one responsible for misogyny identification) was trained using the entire data set, the second CNN (the one responsible for detecting aggressiveness) was trained only on those examples labeled as misogynous.
During testing the same set-up as in Strategy 1 was followed. As such, both CNNs were arranged in a pipeline, with the first one responsible for detecting misogynous tweets, and the second one responsible for identifying aggressiveness, amongst those tweets held misogynous by the first CNN. 5
Results for aggressiveness detection, on the other hand, varied substantially, with the Logistic Regression classifier (Strategy 2) performing worst, when compared to the CNNs used for the same task in the other strategies (7% against Strategy 1, and 18% against Strategy 3).
Interestingly, the CNN trained only on examples labeled as misogynous (Strategy 3) performed better (around 13%) than its counterpart trained over the entire data set (Strategy 1). It is important to recall that this was the only difference between both strategies.
Final results at the competition’s private test set can be seen in Table 3. As it turns out, Strategy 2 was the best ranked of our models, reaching the sixth place at the competition (being only F = 0:03 worse than the winning model). Puzzling enough, this was the model that scored worse in our test set. One possible explanation for this fact might be that our CNN was not capable of generalising over different data sets. Differences in the balance between misogynous and non-misogynous, and between aggressive and non-aggressive examples, in both data sets, might also explain this behaviour. Whatever the reason, we leave this investigation for future work. 6
In this work, we described two models submitted to EVALITA 2020’s subtask A on Automatic Misogyny Identification. To this task, a CNN and an LR classifier were trained, and arranged as a pipeline following three different strategies, with one of them coming at sixth place in the competition.
Even though our classifier turned out to be competitive, we believe improvements could be made to achieve better results, such as the addition of lexical features, for example. Also, it might be that following some preprocessing strategy, such as removing stop words, for example, might result in a better performance.
As for future work, besides testing the above cited changes, it would be interesting investigating why the worst model at the test set (as distributed to all participants) turned out to be the best model at the competition’s private data set. The reasons for this behaviour are something to be determined.