Vendedor virtual com PLN, machine learning e grafos de conhecimento
Carregando...
Data
2021-06-02
Autores
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Goiás
Resumo
The information age, provided by the internet, is revolutionizing the ways in which
services are offered. In this scenario, chatbots, automated response programs in dialog
interfaces, are on the rise. However, it usually only have pre-programmed responses and
the most common options, which can cause discontent in the user. The objective of this
work is a research applied to the design of a chatbot model, with artificial intelligence,
capable of interacting in the user’s natural language, emitting responses like those of a
human within his area of activity, which, in this case, is the sales area. Thus, a chatbot
model was developed and evaluated using the Rasa framework, based on new methods
for obtaining artificial intelligence training data. A Knowledge Graph model has also
been developed that can provide more quality to the service provided, offering products
and understanding details, including the misspellings of messages.
Descrição
Palavras-chave
Chatbot, Rasa, Inteligência artificial, Grafos de conhecimento, Chatbot, Knowledge graphs, Artificial intelligence, Rasa
Citação
SOUZA, Luiz Yokoyama Felix de. Vendedor virtual com PLN, machine learning e grafos de conhecimento. 2021. 72 f. Trabalho de Conclusão de Curso (Graduação) - Escola de Engenharia Elétrica, Mecânica e de Computação, Universidade Federal de Goiás, Goiânia, 2021.