Leveraging Automated Machine Learning for Text Classification: Evaluation of AutoML Tools and Comparison with Human Performance

Recently, Automated Machine Learning (AutoML) has registered increasing\nsuccess with respect to tabular data. However, the question arises whether\nAutoML can also be applied effectively to text classification tasks. This work\ncompares four AutoML tools on 13 different popular datasets, including Kaggle\ncompetitions, and opposes human performance. The results show that the AutoML\ntools perform better than the machine learning community in 4 out of 13 tasks\nand that two stand out.\n