Un outil d’aide à l’annotation en vue de la construction d’un corpus annoté

Abstract

Artificial intelligence plays a crucial role in various fields, revolutionizing industries and enhancing human capabilities. One area where Artificial intelligence has demonstrated immense importance is in natural language processing. It enables computers to understand, interpret, and respond to human language, making it essential for applications such as virtual assistants, language translation, and sentiment analysis. Artificial intelligence powered natural language processing algorithms can analyze vast amounts of textual data, extracting valuable insights and enabling faster decision-making. With its ability to process and comprehend human language, artificial intelligence in natural language processing opens up new possibilities for improved communication, automation, and innovation across multiple domains.Despite its significance, the Arabic language lacks comprehensive research. This hampers progress in natural language processing, machine translation, and language learning. Investing in Arabic language research would refine models, and preserve cultural richness.For this reason, as Arabic speakers, we have chosen the field of Arabic natural language processing. Therefore, the objective of this thesis is to The objective is to utilize deep learning and machine learning techniques to create a semantic roles labeling tool for the Arabic language by training an artificial intelligence model.For our project, we have selected the latest and largest database, OntoNotes 5.0, which is funded and supported by the US Department of Defense, the University of Pennsylvania, the University of Colorado and others.