Team OmarRadi at the Mowjaz Multi-Topic Labelling Task

This paper describes the 7th winning system for the Mowjaz Multi-Topic Labelling Task [1]. The current research work discusses the models used for the experimentation that led to the proposed system and shows the models’ performances on the task’s dataset of Arabic Articles. The proposed system, which was a MLP system with a single hidden layer. It achieved an F1-Sample score of 0.8594, F1-Macro score of 0.8476, F1-Micro score of 0.8504 and an Accuracy score of 82.9 which published on the task’s website which led to rank the model as 7th in the Mowjaz Multi-Topic Labelling Task [1].