A content mining and network analysis method for the thematic evolution of educational leadership

This study presents a novel approach to analyze the thematic evolution in the research area-‘Educational Leadership’. This approach combines the content mining techniques and Network Analysis to detect and visualize the different research thematic topics. The co-word analysis method is used to longitudinally detect the frequency of popular topics and topic words in a given time period (1996-2015). The weighted network structure is used to analyze the relationships between themes through the co-occurrence of themes, with highlighted centrality and community of the network to show the hot topics in the theme and their relationships, and the changes and relationships between themes are presented through dynamic Network Analysis among the 4 periods (1996~2000, 2000~2005, 2006~2010, 2010~2015). With highlighted network centrality & community to show the hot topics in the theme and their relationship, and dynamic Network Analysis to present the changes and trends among the 4 periods (1996~2000, 2000~2005, 2006~2010, 2010~2015). Compared with traditional methods, our method provides direct visual effects, thereby making the structure scalable and providing easy-to-understand methods for public end users. The method in the current study can be widely used in other research fields.