Construction and visualization of a knowledge graph based on student consumption data

Abstract

The research fields of knowledge graphs and data mining have received widespread attention in recent years because these technologies not only provide users with a comprehensive and easy-to-use knowledge experience by linking data as entities and relationships, but also can support complex queries and analyses, thereby enabling Information retrieval is more efficient.

In order to understand the consumption situation and consumption habits of students, this article conducts an in-depth analysis of the consumption data of students within the university. Students are divided into four consumer groups using a clustering algorithm such as K-means. There are significant differences between these four groups in terms of average monthly consumption amount and the rate of eating three meals. Based on the analysis results, entities and their interrelationships are constructed and designed to form a knowledge graph. The knowledge graph includes four entities: students, consumption orders, shops, and consumption level clustering categories, with a total of 64,829 entity nodes created. Through these data relationships, students' consumption behavior can be intuitively displayed.

Data mining technology provides powerful tools for educational intelligent management to predict future trends and optimize strategies. Relevant school departments can use these data as a basis for decision-making and reveal the hidden information behind student consumption data.