Frequency detection of experimental errors through Learning Analytics techniques

The process that systematically collects and analyzes large volumes of data in order to improve the teaching-learning process is called Learning Analytics. Based on data processing, educational data mining and visualization, it is possible to monitor academic progress, enhancing actions on how the teacher should conduct the discipline. The objective of this research is to mine data from experiments carried out in the remote laboratory called “Virtual Instrument Systems in Reality”. In order to create classification groups according to theoretical analysis studied in circuit analysis. The k-NN classification model was used for this research. The algorithm presented a very satisfactory result, its accuracy resulted in samples with values greater than 0.9. Considering an excellent form of classification analysis for circuits with 1: 1, 0: 1, 1:0 model.