PROJECT TITLE :

Mining Online Discussion Data for Understanding Teachers' Reflective Thinking - 2017

ABSTRACT:

Teachers’ online discussion text knowledge streamline their reflective thinking. With the growing scale of text information, the ancient method of manual coding, but, has been challenged. So as to method the massive-scale unstructured text data, it's necessary to integrate the inductive content analysis technique and academic data mining techniques. An inductive content analysis on samples taken from 17624 posts was implemented and the classes of lecturers’ reflective thinking were obtained. Based mostly on the results of inductive content analysis, we have a tendency to implemented a single-label text classification algorithm to classify the sample information. Then we applied the trained classification model on a large-scale and unexplored online discussion text information set and two sorts of visualizations of the results were provided. By using the classes gained from inductive content analysis to create a radar map, teachers’ reflection level was represented. Additionally, a cumulative adjacency matrix was created to characterize the evolution of teachers’ reflective thinking. This study may partly make a case for how lecturers mirrored in on-line professional learning environments and brought awareness to academic policy manufacturers, teacher coaching managers, and education researchers.