Deep Learning-Based Digital Assistant for Farmers in South Africa

Abstract

A digital assistant is a sophisticated dialogue software that transcends the use of Machine Learning, Artificial Intelligence (AI), and Natural Language Processing (NLP). The main goal of a digital assistant is to assist humans with different tasks depending on their domain. Many of these digital assistants use only English as a means of communication. For many countries, like South Africa, that are multi-lingual this causes a problem, especially for monolingual people in rural areas, as most use their native languages to communicate. This paper proposes a Deep Learning-based digital assistant for rural and urban or commercial farmers in South Africa. The proposed system uses two South African languages, namely English and Zulu. The digital assistant uses Natural Language and Deep Learning algorithms to increase the accuracy of the responses. These results show that the model has a 0.9648 accuracy.