**2.3 Online assessment trends**

Assessment methods, such as coursework, examinations, and viva are key to evaluating students' learning. Traditionally, the assessment was tasked to teachers, who allocated a particular grade based on their judgement and/or set marking criteria. Nowadays, with increasing reliance on online education, there is a growing need for new innovative technologies to assess students' learning. AI-based technologies and other semantic technologies, such as Natural Language Processing (NLP), have been identified as being effective solutions in this aspect. Machine learning (ML) techniques, involving trained algorithms, can help teachers in complex assessments as well as reducing the burdens of human marking, time, and cost.

Many studies focusing on the use of ML techniques for students' assessment, relied on the validity of the work while undermining the technical and pedagogic features in evaluating student works in science subjects, using approaches such as text recognition, classification, and scoring [17]. Disagreements related to vocabulary between human and ML scoring are possible when assessing students' works [18]. To address this, NLP techniques can be used in the assessment process. NLP algorithms have been found to be effective in assessing students' work, such as essays, marking closely the same as humans. Moreover, they can provide similar scores in both original languages and translated versions of essays, thus reflecting the applicability of assessment technologies across multiple languages. NLP techniques have been shown to provide consistency, scalability, and traceability in the form of an automated marking system. Furthermore, online exam invigilation proctoring techniques, such as audio/video/screen forms, automated AI invigilation, preventing HDMI cable extensions, avoiding background screen sharing, preventing unauthorized access, are a few of the new techniques being used in online exams and assessments aimed at avoiding cheating and/or copying. In addition, AI-based solutions, such as Expert Control System (ECS)-based tutoring platform and Agent-based Tutoring Systems (AbS) are proving to be effective in assessment for learning [19, 20].

### **2.4 Self-paced learning**

Each student has a different set of learning-related abilities, strengths, and weaknesses. While students like a particular subject, they may dislike others and, as result, can be weak in these. To overcome this, self-paced learning, where students study at their own pace, with little influence from classroom lectures, can be one of the effective solutions. Technology plays an important role in driving self-paced learning. Many universities have launched online courses that give the students more control over their study, so they can study at their own pace. Blended learning, where significant elements of the learning environment, such as face-to-face online tools, are used in learning, can also support students' self-paced learning and enhance their engagement in learning [21]. Also, empirical findings have suggested that self-learning tools, with the support of relevant pedagogy and learning processes such as self-regulated learning can significantly improve students' learning and engagement [22]. In sum, online educational tools allow for self-paced learning and this can significantly improve students' engagement in learning and knowledge retention.

### **2.5 Artificial intelligence and machine learning**

AI is already being implemented in various business sectors. The education sector has a range of areas in which AI can be of significant help in improving the *Embracing Technological Change in Higher Education DOI: http://dx.doi.org/10.5772/intechopen.100431*

learning process. For instance, AI-based auto-invigilation can be used for administering online examinations. Using video, a remote invigilator can watch candidates during an exam, while audio invigilation can capture sound coming from candidates' backgrounds while taking the exam (recording any malpractices such as cheating). Moreover, facial recognition and biometrics can provide added security for verifying the candidates before accessing various online resources. However, using AI techniques in certain areas, including invigilation, is an issue that has been subject to debate with some arguing that it is unethical. Furthermore, data-driven analysis using AI and ML techniques can enhance decision-making capabilities in the education sector. For example, in learning analytics, historical school dropout data can be used to train ML algorithms to predict future dropouts based on numerous variables including dropout rates by class, level, gender, region (urban/rural), and college type. This can help institutions and governments in taking effective decisions aimed at reducing student attrition. The reliability of AI techniques has been proved to be effective in various studies [23]. For instance, a recent work [24] involved investigating the use of AI-based algorithms in the admissions process at a German university, where it was found that the decisions were more effective than those of humans. This is clear evidence that AI-based solutions can improve various operations in educational institutions.
