文献类型：期刊文献

英文题名：A Framework for the Design of Plant Science Education System for China’s Botanical Gardens with Artificial Intelligence

作者：Guo, Lijuan[1]; Wang, Jiping[1]

机构：[1] School of Art Design and Media, East China University of Science and Technology, Shanghai, China

年份：2020

卷号：1294

起止页码：267

外文期刊名：Communications in Computer and Information Science

收录：EI(收录号：20205009597572)

语种：英文

外文关键词：Learning systems - Environmental protection - Plants (botany) - Deep learning

摘要：The framework of the traditional plant science education has lagged, which is hard to meet the needs of various visitors in China. In the following paper, we propose a unified interaction framework to help the design of the plant science education system. That is suitable for various visitors from China. They are from young to old with different levels of education, and some of them have dialect or accents. Inspired by the background, we try to design a new framework that takes advantage of the variety of the visitors’ output, aiming to make the plant science education system very smart to understand what the visitors want to learn. Firstly, all the plants are numbered with the non-linear digital numbering method, which can be seen as the label of the training process for deep learning networks. Afterward, the interaction continuously collects the visitors’ plant-related voice data as the input of the deep learning networks during the operation of the system to improve the performance and the stability of the plant science education system. Through continuous training, the overall accuracy of the system could be improved, and the system can gradually understand the regular pattern and central issue of people cares for the plant. The framework provides a new idea for the science education of the botanical garden and further improves the level of science education in China. This framework helps achieve sustainable development and environmental protection. ? 2020, Springer Nature Switzerland AG.