The detection of hate speech has become a priority, given the vast amount of information that is spread every day on the Web. These communications are commonly targeted to a group of individuals with a specific ethnicity, gender, sexual orientation, or religion. Different key factors have influenced the growth of this threat, like social polarization, virality, and anonymity. To build a healthy society and fight against racism, misogyny, xenophobia, or another shape of hate is fundamental to detect these comments.
Within this context, this project aims to detect hateful comments addressed to the leading female political leaders in Spain by developing automatic tools based on machine learning and natural language processing.
Key words: hate speech, hateful comments, political leaders, machine learning, natural language preprocessing.
- Task 1. Analyze corpus.
- Task 2. Descriptive analysis.
- Task 3. Discuss labels and label data? & Define project goals.
- Task 4. Build baseline models.
- Task 5. Build sophisticate models.
- Task 6. Evaluate models.
- Task 7. Conclusions.