Abstract
This paper presents a novel Arabic Conversational Intelligent Tutoring System (CITS) that adapts the learning styles VAK for autistic children to enhance their learning. The proposed CITS architecture uses a combination of Arabic Pattern Matching and Arabic Short Text Similarity to extract the responses from the resources. The new Arabic CITS, known as LANA, is aimed at children with autism (10 to 16 years old) who have reached a basic competency with the mechanics of Arabic writing. This paper describes the architecture of LANA and its components. The experimental methodology is explained in order to conduct a pilot study in future.