EXIST-2021: sEXism Identification in Social neTworks

Detecting online sexism may be difficult, as it may be expressed in very different forms. The aim of this competition and project is the detection of sexism in a broad sense, from explicit misogyny to other subtle expressions that involve implicit sexist behaviours. The automatic identification of sexisms in a broad sense may help to create, design and determine the evolution of new equality policies, as well as encourage better behaviors in society.

I applied fine-tuning to bert-base-multilingual-uncased to make the classification. I chose this model because the task contained text in English and Spanish. In the table below you can see the performance of my model, which is quite close to the winner of the shared task (see EXIST-2021 results).

The code with more explanations is in my GitHub EXIST2021.