TY - CONF TI - An Analysis of ResNet50 Model and RMSprop Optimizer for Education Platform Using an Intelligent Chatbot System AU - Saadna, Youness AU - Boudhir, Anouar Abdelhakim AU - Ben Ahmed, Mohamed A2 - Ben Ahmed, Mohamed A2 - Teodorescu, Horia-Nicolai L. A2 - Mazri, Tomader A2 - Subashini, Parthasarathy A2 - Boudhir, Anouar Abdelhakim T3 - Smart Innovation, Systems and Technologies AB - A chatbot is a software (or machine) that has the ability to talks with a user: it is a virtual assistant that can answer a number of user questions, and providing the correct responses. In the last few years, the use of chatbots is very popular in various fields, such as health care, marketing, educational, supporting systems, cultural heritage, entertainment, and many others. This paper proposes an intelligent chatbot system that can give a response in the form of natural language or audio to a natural language question or image input in different domains of education and will support multiple languages (English, French, and Arabic). To realize this System, we used different deep learning architectures (CNN, LSTM, Transformers), computer vision, transfer learning to extract image features vector, and natural language processing techniques. In the end, after the implementation of the proposed model, a comparative study was conducted in order to prove the performance of this system using image-response model and question-response model using accuracy and BLEU score metrics. C1 - Singapore C3 - Networking, Intelligent Systems and Security DA - 2022/// PY - 2022 DO - 10.1007/978-981-16-3637-0_41 DP - Springer Link SP - 577 EP - 590 LA - en PB - Springer SN - 9789811636370 KW - Artificial intelligence KW - Chatbot KW - Education ER -