Visual Mining Research On Student Science Literacy Evaluation Data

It is generally accepted internationally that scientific literacy refers to the ability to use scientific knowledge to identify problems and make evidence-based conclusions in order to understand and make decisions about the natural world and the changes in the natural world through human activities.As a basic goal of international science education,Scientific Literacy is the premise and foundation of the two goals of popularizing scientific knowledge and improving education quality in the current scientific education reform.The study of scientific literacy is inseparable from the assessment of the level of scientific literacy.The evaluation and training of students' scientific literacy level is of fundamental significance to the long-term development of various countries and the improvement of national quality.Experts in the field of science education research in China have combined information technology,network technology and sensor technology to build evaluation programs,and build a corresponding scientific literacy evaluation cloud platform.The platform has been put into preliminary practical evaluation work.However,with the deepening of scientific literacy evaluation research,the scale of data collection is increasing.How to better interpret the evaluation data and enable researchers,teachers and families in the field to obtain the hidden value behind the data more quickly,it is necessary to consider introducing new technologies and methods.This research applies data mining and data visualization technology to the analysis of scientific literacy evaluation results,and obtains the relationship between the results of the results based on fuzzy clustering algorithm and the dimensions of scientific literacy evaluation based on association rules algorithm.In the data visualization,based on the front-end technology,design and implement the data visualization platform,using a variety of chart forms such as ring graph,radar map,scatter plot,etc.and combine with various aspects of visualization,from the evaluation of each attribute value of the data.Observing the rules,obtaining valuable information from the law,and embedding the fuzzy C-means clustering algorithm for performance rating into the platform to facilitate future evaluation.