Comparative Study on Green Logistics Based on Neural Network Word Embedding

• 1、School of Management，Wuhan University of Science and Technology，Wuhan，430065

Abstract：In the process of sustainable development path, China\'s logistics industry is actively responding to and promoting the development of green logistics. Green logistics has received wide attention in the academic and practical circles. This research combines academic platforms and media platforms, and uses neural network word2vec word embedding algorithm to carry out comparative research on green logistics by word embedding algorithm, aiming to promote the sustainable development of green logistics. First of all, CNKI and China Logistics and Purchasing network were selected as the objects of comparative research, and the corpus of "green logistics" was collected based on these two platforms respectively, thus building two corpura of academic platforms and media platforms. Next, word2vec was used to model the two corpora separately. On this basis, the top 20 words that were most similar to "logistics" and the top 20 words that were most similar to "green logistics" are combined to compare the hot spots of the two platforms. Then, the t-SNE algorithm was used to reduce the dimension of the word vector, and the matplotlib.pyplot package. The results show that green logistics varies between academic and media platforms. This study innovatively uses the word2vec algorithm to compare the differences in the semantic aspects of green logistics research interests and word embedding profiles between academic platforms and media platforms, which provides a new perspective for the research of green logistics.

Keywords： green logistics comparative research word2vec academic platforms media platforms