The Impact of Data Pre-Processing on Hate Speech Detection in a Mix of English and Hindi–English (Code-Mixed) Tweets (MDPI)

Due to the informal format of tweets with variations in spelling and grammar, hate speech detection is especially challenging in code-mixed text. In this paper, we tackle the critical issue of hate speech detection on social media, with a focus on a mix of English and Hindi–English (code-mixed) text messages on Twitter. More specifically, we aim to evaluate the impact of data pre-processing on hate speech detection. Our method first performs 10-step data cleansing; then, it builds a detection method based on two architectures, namely a convolutional neural network (CNN) and a combination of CNN and long short-term Memory (LSTM) algorithms. We tune the hyperparameters of the proposed model architectures and conduct extensive experimental analysis on real-life tweets to evaluate the performance of the models in terms of accuracy, efficiency, and scalability. Moreover, we compare our method with a closely related hate speech detection method from the literature. The experimental results suggest that our method results in an improved accuracy and a significantly improved runtime. Among our best-performing models, CNN-LSTM improved accuracy by nearly 2% and decreased the runtime by almost half.

https://www.mdpi.com/2076-3417/13/19/11104