Publicado en 3C Empresa – Volume 13, Issue 1 (Ed. 53)
Autores
Resumen
Abstract
In order to explore the connection between food culture and regional characteristics in ancient literature, a text clustering model is used to bring together literary works with similar food culture descriptions. First, selection criteria such as era, region, and author's social background are set. Then, the works are iteratively assigned to the closest cluster centers by the K-Means algorithm, and these cluster centers are continuously updated to find the best clustering results. In the text preprocessing stage, keywords related to food culture were extracted from each work to form a basic feature set. Finally, the K-Means algorithm is used to identify food culture themes with different regional characteristics. The entropy values of the text clustering model are 92.85 and 72.6, which reveal the common dietary elements of the literary works in each cluster and reflect the close relationship between regional characteristics and dietary habits.
Artículo
Palabras clave
Keywords
Text clustering; K-Means algorithm; cluster center; text preprocessing; food culture
Articulos relacionados