Publicação

Ai-driven decision support in the automotive industry: designing a user-centric ai chatbot using large language models and the double diamond approach

Ver documento

Detalhes bibliográficos
Resumo:Companies are confronted with a paradoxical situation: although data is abundant, its volume complicates rather than facilitates decision-making. This thesis explores how large language models (LLMs) can address this challenge with a user-centered approach. Based on the specific needs and challenges of decision-makers in the automotive industry, identified through user interviews and content analysis, a chatbot was developed and evaluated. The findings demonstrate the potential of LLMs to streamline decision-making by efficiently processing complex data and generating insights. This research showcases the feasibility of user-centered AI tools in enhancing decision-making processes and provides a comprehensive framework for future research.
Autores principais:Preiss, Florian Fritz
Assunto:Ai-driven decision-making User-centric design Chatbot development Large language models Prompt engineering Retrieval-augmented generation
Ano:2024
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso embargado
Instituição associada:Universidade Nova de Lisboa
Idioma:inglês
Origem:Repositório Institucional da UNL
Descrição
Resumo:Companies are confronted with a paradoxical situation: although data is abundant, its volume complicates rather than facilitates decision-making. This thesis explores how large language models (LLMs) can address this challenge with a user-centered approach. Based on the specific needs and challenges of decision-makers in the automotive industry, identified through user interviews and content analysis, a chatbot was developed and evaluated. The findings demonstrate the potential of LLMs to streamline decision-making by efficiently processing complex data and generating insights. This research showcases the feasibility of user-centered AI tools in enhancing decision-making processes and provides a comprehensive framework for future research.

Atividades financiadas

Não foram encontrados projetos financiados.