Eğitim alanında veri madenciliği ve bilgi keşfi

This thesis aims to increase knowledge discovery in the field of education by developing data mining methods. Considering that there is a large amount of ordinal and unlabeled data in the education area, this study focused on using semi-supervised classification and ordinal classification techniques. Semi-supervised learning is a type of machine learning technique that constructs a classifier by learning from a small collection of labeled samples and a large collection of unlabeled ones. Although some progress has been made in this research area, the existing semi-supervised methods provide a nominal classification task. However, semi-supervised learning for ordinal classification is yet to be explored. To bridge the gap, two concepts, "semi-supervised learning" and "ordinal classification", were combined in this study for the categorical class labels for the first time and introduced a new concept of "semi-supervised ordinal classification". Our study proposes a new method for semi-supervised learning that takes into account the relationships between the class labels, especially class orderings such as low, medium, and high. We performed an extensive empirical study that involved 10 benchmarks and 3 educational ordinal different quantities of labeled datasets with samples varying from 15% to 50% with an increment of 5%, aiming to evaluate the performance of our method by combining different base learners. The experiments showed that the proposed method improved the classification accuracy of the model compared to the existing semi-supervised method on ordinal data. We also developed a web application to provide the accessibility of our method.