Nowadays, educational institutions gather and save a vast amount of data, including records of student enrollment, attendance, and exam outcomes. Such data can be mined to produce interesting information that is helpful to those who handle it. New ways to assess e-pedagogical learning's efficiency are required due to the increased interest in it as a crucial teaching and learning process. This paper offers a data mining research scenario for the next 20 years where the setting is e-learning, and the key topic is the assessment component, which is seen as a latent issue in this environment. Our objective is to present an understudied review of educational data mining research on the teaching and learning process while taking the viewpoint of education into consideration.

ion The rise of new areas of information technology and the ever-increasing significance of interactive features in e-learning environments have inspired adaptive e-learning. Although researchers have been working incessantly for the past 20 years to include intelligent and adaptive elements in e-learning software, these features are typically ad hoc and only serve partial views. Due to the lack of a unified structure, framework, or standard, many adaptive e-learning perspectives frequently have characteristics that are repeated and overlapped with one another. The main elements of adaptive e-learning are identified in this paper, along with their unique viewpoints. We present a succinct overview of the approaches in use for the various adaptive aspects of these components. Additionally, we also present a hierarchical structure of several adaptive e-learning components, which serves as a framework for this review study.