Chatbot como ferramenta para auxiliar pequenos produtores na escolha da espécie forrageira

• Ano de defesa: 2024
• Autor(a) principal: Oliveira Filho, Aloísio Alkmim de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Agronomia; UFSM; Programa de Pós-Graduação em Agricultura de Precisão; Colégio Politécnico da UFSM
• Palavras-chave em Português: Chatbot; inteligência artificial; Pastagem; Artificial intelligence; Pasture; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA
• Link de acesso: http://repositorio.ufsm.br/handle/1/34404

Ruminant production in Brazil is predominantly extensive. This is feasible due to the vast expanse of land in the country and the fact that it is a more economical form of production compared to confinement systems. Additionally, pastures, which provide nutritional support for the animals, significantly contribute to reducing production costs. However, it is well known that there is no miraculous forage plant that, regardless of the property's edaphoclimatic conditions, offers high productivity and good acceptability by the animals. Nonetheless, there is a multitude of forage plants that can be implemented in different locations, each with its specific characteristics that must be respected. Consequently, small-scale livestock farmers often face difficulties in making the best decision regarding the choice of the most suitable forage plant for their region. This study presents a tool developed to assist rural producers in making decisions about which forage plant is most appropriate for pasture implementation in their locality. This tool was designed as a chatbot, with the aim of being a user-friendly system, utilizing Google's artificial intelligence platform, Dialogflow. This platform is integrated with major messaging applications, such as WhatsApp and Telegram, facilitating its adoption since these apps are widely used by the Brazilian population. The tool operates by filtering the information provided by the livestock farmer. This information is collected through questions posed by the chatbot to the user. The questions are formulated in an accessible manner, enabling individuals with varying levels of education to easily understand and interact with the system. The questions address issues related to the region of the country where the property is located, the type of soil on the property, regional rainfall, soil fertility, among other factors. At the end of the process, the system delivers a report with the most suitable forage plants for that locality, highlighting the pros and cons of each, thus facilitating the rural producer's decision-making process. The forage plants included in the system's library were selected based on their ease of availability in stores specializing in seed sales, ensuring that the livestock farmer can find them without difficulties in their locality, avoiding frustration. Additionally, there is the possibility to add or remove forage plants from the system's library as needed. The development of the chatbot is the first stage of the project; subsequent stages involve usability testing and field analysis with livestock farmers. Thus, this system represents a significant advancement in the application of artificial intelligence to agriculture, specifically in assisting small-scale farmers with precision farming techniques. By simplifying the forage plant selection process, this chatbot has the potential to increase productivity and sustainability in ruminant production across various regions of Brazil.