Implementation of specialized methods of educational data mining for training future vocational education teachers

Abstract

The article presents a detailed analysis of modern methods of Educational Data Mining (EDM). Based on this analysis, an integrated set of approaches has been developed, aimed at optimizing the process of training educators in professional education. The article describes key EDM techniques, including prediction, clustering, relationship mining, data distillation for human judgment, and discovery of new knowledge through models.

Educational data mining methods have been enhanced with specialized techniques for pre-processing digital images and artificial intelligence algorithms, tailored to meet the educational needs of future vocational education teachers. The utilized methods of digital image pre-processing, including filtering, contrast enhancement, and contour detection, improve the quality and accuracy of image processing, facilitating more effective analysis in the context of training future vocational education teachers.

In the context of artificial intelligence implementation, image segmentation and object detection techniques using fuzzy logic have been applied to enhance accuracy and adaptability. Among the methods used, it is worth noting the application of convolutional neural networks (CNNs), which provide effective object detection in images, as well as the Viola-Jones method, known for its ability to detect objects quickly and accurately. These technologies significantly enhance the potential of pedagogical training, as they allow for the development of intelligent systems capable of adaptively processing visual information and automating analytical processes in professional education.

Methods for integrating fuzzy logic with core analytical processes such as data clustering, image segmentation, and object detection were thoroughly examined. The application of fuzzy logic facilitates effective consideration of uncertainties in clustering, enhances the accuracy in identifying objects and contours in images, and boosts the efficiency of object detection under complex visual conditions. This approach improves the processing and analysis of visual data, which is crucial in the training of future vocational education teachers.