Benefits of using data mining techniques to extract and analyze Twitter data for higher education applications: a systematic literature review

In recent years, there has been a growing interest by education actors to include TIC in their institutions; as well as social networks, far from being a problem and their use aimed, permit innovate traditional classes and improve communication between teachers and students This study has two objectives: (1) conduct a systematic literature review through searching papers published between January/2007 and March/2019 in data bases like as ACM, IEEE, ScienceDirect, Springer and others, to evidence researches that apply data mining techniques to extract and analyze Twitters data in higher education; and (2) to emphasize pedagogic practices that include Twitter and data mining to improve education process. From 315 papers obtained, only 65 fulfilled inclusion criteria. The main results indicate that: (1) the most used data mining techniques are predictive with classification tasks; (2) Twitter is principally used to: (a) determinate perception; (b) share information, materials and resources; (c) generate communication and participation; (d) promote abilities and (e) improve oral expression and academic performance; (3) United States has the most numbers of researches in this area; however, in Latin-American countries findings are not enough, so, there a new area to investigate in this region and (4) researches used models, methods, strategies, theories and instruments as a pedagogic practice; so that, there wasn't an agreement about a shape to include Twitter data extracting in higher education to improve teaching and learning process.