Unstructured data in social media is as part of the “big data” spectrum. Unstructured data in Social media can provide useful insights into social phenomena and citizen opinions, both of which are critical to government policy and businesses decisions. Teachers of business intelligence and analytics commonly use quantitative data from sales, marketing, finance and manufacturing to demonstrate various analytics concepts in a business context. However, researchers have seldom used social media data to analyze social behavior and communication. In this study we aim to demonstrate an assessment structure for teaching social media analytics concepts with the goal of analyzing and interpreting social media content. We base this assessment on forum postings regarding two recent events: the Christchurch earthquake in New Zealand, and the Japanese earthquake and tsunami. The aim of the assessment is to discover social insights. We base the assessment structure on Cooper’s Analytics Framework to cover such concepts as term frequency (TF), term frequency–inverse document frequency (TFIDF), data visualization, sentiments and opinions analysis, the Nearest Neighbor K-NN classification algorithm, and Information Diffusion theory. We review how the students performed on the assignment that used this assessment, and we make recommendations for future studies.
Goh, Tiong T. and Sun, Pei-Chen
"Teaching Social Media Analytics: An Assessment Based on Natural Disaster Postings,"
Journal of Information Systems Education: Vol. 26
, Article 3.
Available at: https://aisel.aisnet.org/jise/vol26/iss1/3