A retail conversational interface to naturally capture unstructured information triggering high value analytical and operational business actions

• Autor(a) principal: Rui Miguel da Silva Gomes
• Data de Publicação: 2017
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/10216/106502
• topic: Engenharia electrotécnica, electrónica e informática; Electrical engineering, Electronic engineering, Information engineering

The retail work environment is hectic at times and there is little to no time to write something down or use the appropriate processes to pass information down to the right recipient. This lack of streamlined processes causes information to be lost or processed slowly, which is troublesome for retail businesses. Actions that could be taken instantaneously can take up to several weeks to be processed, and non-critical but important data can be lost, which can definitely be improved by using a better process. The recent breakthroughs in modern voice recognition technologies and natural language processing can be adopted to create a voice user interface that is able to process spoken commands, providing an overall better experience for retail businesses, trigger customizable actions in real time, help in inventory replenishment, expedite maintenance operations and increase customer satisfaction levels. The data input step goes from a slow and cumbersome one, such as writing down a note, to just speaking to a system that is intelligent enough to not only convert the user's voice to text, but also understand the meaning of said words. That data can then be stored and analysed, empowering retail businesses by providing high value analytical information and triggering operational business actions. These actions can be customized to fit the needs of different businesses.