Publicação
Development of a chatbot to support an university institutional website
| Resumo: | Currently, the institutional website of the Polytechnic Institute of Braganc¸a (IPB) lacks an interactive system that allows users to obtain information quickly and efficiently. With the increase in the use of online services and the preference of students for instant interactions, there is a growing need for tools that facilitate communication and access to information. This work proposes the development of a chatbot to support the IPB website, with the aim of improving user experience and providing quick and accurate answers to their queries. The main objective of this project is to create an inteligente chatbot capable of understanding and responding to a wide range of questions related to IPB. To achieve this, modern natural language processing and machine learning technologies such as Streamlit, Langchain, and OpenAI are used. The system includes an automated Web Scraper to keep information up-to-date, an architecture based on efficient document segmentation, and an administrative back office for monitoring and management. This development represents a significant advance in how users interact with the IPB website, offering a more efficient and accessible communication channel. The chatbot has the potential to reduce the workload of administrative staff while improving user satisfaction by providing accurate and instant answers to their questions. |
|---|---|
| Autores principais: | Sampaio, Telmo |
| Outros Autores: | Oliveira, Pedro Filipe; Matos, Paulo |
| Assunto: | Chatbot Webscraper Machine-learning Streamlit |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | comunicação em conferência |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Instituto Politécnico de Bragança |
| Idioma: | inglês |
| Origem: | Biblioteca Digital do IPB |
| Resumo: | Currently, the institutional website of the Polytechnic Institute of Braganc¸a (IPB) lacks an interactive system that allows users to obtain information quickly and efficiently. With the increase in the use of online services and the preference of students for instant interactions, there is a growing need for tools that facilitate communication and access to information. This work proposes the development of a chatbot to support the IPB website, with the aim of improving user experience and providing quick and accurate answers to their queries. The main objective of this project is to create an inteligente chatbot capable of understanding and responding to a wide range of questions related to IPB. To achieve this, modern natural language processing and machine learning technologies such as Streamlit, Langchain, and OpenAI are used. The system includes an automated Web Scraper to keep information up-to-date, an architecture based on efficient document segmentation, and an administrative back office for monitoring and management. This development represents a significant advance in how users interact with the IPB website, offering a more efficient and accessible communication channel. The chatbot has the potential to reduce the workload of administrative staff while improving user satisfaction by providing accurate and instant answers to their questions. |
|---|