Multimodal Interaction Techniques for Chatbots

• Autor(a) principal: Fornelos, Tiago Filipe Modesto
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/145194
• topic: Chatbo; Multimodal Interaction; User Interface; E-commerce; 3D Environments; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

Chatbots have become popular and are being used in the most diverse areas. The most common use for chatbot are the ones integrated in our smartphones, like Google Assistant and Siri. This kind of technology evolved significantly in the past few years and now the users have the feeling to be talking to another human being. In a near future, most of the interaction between large organisations and their users will be mediated by AI agents. A market that benefited a lot from this arising technology are the e-commerce websites. Sometimes, it can be a struggle to find a product in a web site due to bad interface designs. When integrated in these websites, the clients have immediate access to a seller that is able to help them find the product they want or clarify some doubts that they might have. Despite the great evolution in the responsiveness of these systems, when talking about the User Interface (UI) of the systems, there are no major changes. Normally, chatbots are represented using the common chat box in the bottom right corner of the screen. In the scope of iFetch project, that the main focus is revolutionise the chatbot assistance in online sales, it is proposed for this dissertation the development of a user interface for a chatbot with multimodal interaction, making the user experience more interactive and engaging. This interface is a representation of a virtual room where the user can see the products that he/she wants to buy. The is also a component of image processing to be able to use the 2D images for the generation of their 3D representation.