Can social data mining reveal student feelings? A new study in the International Journal of Computer Applications in Technology suggests that this could be the case given that a student’s digital footprints on social media often reveal their personal experiences and opinions on a variety of subjects, including their course, their friends and family, and their own mental health.
Hua Zhao, Yang Zuo, Chunming Xu, and Hengzhong Li of the College of Computer Science and Engineering at Shandong University of Science and Technology in Qingdao, Shandong, China, suggest that educators and course administrators could benefit from social data mining to understand their students’ moods and provide appropriate help when needed. The same social data mining might also be used to spot trends and patterns as a course progresses and perhaps allow the course itself to be adapted within limits to best serve the students and their education.
Of course, there are such vast amounts of social data online and more added each data that mining such information can only ever tap the seam rather than extracting all of the putative knowledge within. At least that is the received wisdom, but the development of novel data mining tools and artificial intelligence algorithms might change that allowing new insights to be extracted in a timely manner.
The team has developed a new approach to help them understand the emotions and moods of a sample of Chinese students. First, they collect the appropriate social data related to the students and then build a hierarchy category system based on a content analysis. In the second step, they apply an effective multi-class classification method to classify the data into several categories of concern. Finally, a “sentiment” analysis around each category is undertaken to look for emotional content and language that can reveal changing moods in the students as a group or individually, for instance, surrounding exams and other matters. Obviously, exams are a major concern of students the world over and such an approach might be applicable elsewhere in social data mining student mood.
several categories of concern. Finally, a “sentiment” analysis around each category is undertaken to look for emotional content and language that can reveal changing mood in the students as a group or individually, for instance, surrounding exams and other matters. Obviously, exams are a major concern of students the world over and such an approach might be applicable elsewhere in social data mining student mood.
Zhao, H., Zuo, Y., Xu, C. and Li, H. (2021) ‘What are students thinking and feeling? Understanding them from social data mining’, Int. J. Computer Applications in Technology, Vol. 65, No. 2, pp.110–117.