Cluster Analysis and Theme Deconstruction of the Project of Columbia Center for Oral History postprint

Abstract: [Purpose/significance] To explore the characteristics of Columbia Center for Oral History (CCOH) project and deconstruct the theme of the project from a new perspective, so as to provide a reference for enriching the theme of oral history project in China.[Method/process] 9 586 oral history projects of CCOH were selected as the research objects, data were captured by Python, and 3 589 projects (5 997 projects without brief introduction) were filtered through Word2Vec and K-means, and the main theme of the projects was deconstructed.[Result/conclusion] The project of CCOH consists of four thematic levels:①Taking people as the theme - all inclusive, radiating to all walks of life;②taking institutions as the theme - learning from all strengths, giving full play to their own functional advantages;③taking regions as the theme - overall planning, linkage of regional system construction;④taking specific periods, specific movements / events as the theme - innovation paradigm, reflecting on the theme of home and country times.