Big data analysis based on student activity analysis using enhanced decision tree vector quantization

Big data analysis is a learning management system of how the education sector of mining and learning analysis of education data has been proven to be effective and has been widely used in applications which are currently under development. Big data analysis is applied in the higher education institutions for the importance of e-learning and to investigate the use of e-learning as a basis for big data and its “influence”. The online content is to be used to derive some of the data from the student activities and big data. The previous methods could not classify the results of student activity accuracies, learning performance and student educational ranking scores; higher education institutions operate in an increasingly complex and competitive environment. To overcome the above issues, the new method Enhanced Decision Tree Vector Quantization (EDTVQ) is introduced based on Machine learning which improves the imbalance dataset and identifies the contemporary challenges faced in higher education institutions in terms of student activities and explores the potential of big data in overcoming these challenges. Initially, Preprocessing is done using data processing steps for identifying and handling the missing values using Ensemble Principal Components Analysis (EPCA). Next, extracting the learning features from the preprocessed data is done and then the datasets are trained and tested for classification for better results. A student academic activity-based ranking score analysis model is implemented to evaluate the learner’s activities and exam score. This proposed method analysis results evaluate the following performance matrices namely sensitivity, specificity, error rate, classification accuracy and time complexity. In this simulation analysis, it is confirmed that the proposed method EDTVprovides efficient classification accuracy with less error rate and less time complexity compared to other previous methods.