Artificial Intelligence and Machine Learning in Higher Education

Artificial intelligence and machine learning have the potential to address many of the problems that have emerged in higher education due the rapid and haphazard transition to online learning brought about by the coronavirus pandemic. These problems include students' struggle to self-regulate their learning, the increase in curriculum planning and administrative workload for teachers, and the loss of personalized interaction between students and teachers. This chapter explores how artificial intelligence can be used to help students and teachers to adapt to the new realities of online learning, and how these technologies could further transform higher education in the future. By providing more personalized, flexible, inclusive, and engaging learning experiences, artificial intelligence has the potential to re-invigorate students and teachers both and to make virtual classrooms more meaningful and productive.

Pedagogical Ontology Modelling for Cell Biology Domain With an Algorithm for Question Generation

In this chapter, an ontology that structures all the cell organelles and their parts are modelled to cognitively model domain knowledge by explicitly establishing relationships among them. The ontologies are modelled depicting the cell as a system and the parts of the cell as the subclasses of the cell along with various functionalities and behavior. The model further focuses on education pedagogy to generate questions based on the modelled ontologies. Furthermore, the defined ontologies are made consistent by defining the classes and the relationship between them, initializing the instances and axiomatizing the developed ontological content. The modelled ontologies are semiotically evaluated using various learners and domain experts. An overall reuse ratio of 0.91 has been achieved, and the proposed ontology has been differentiated from the existing cell ontologies by focusing on an educational pedagogy. Ultimately, an ontology-focused algorithm for multiple choice question generation has been proposed for cell biology as a domain of choice with an accuracy of 90.03%.

Applications of Machine Learning in Education

This research aims to explore how to enhance student engagement in higher education institutions using novel chatbots. This study's principal research methodology is design science research, which is executed in three iterations: personas elicitation, a survey and development of student engagement factor models (SEFMs), and chatbot interaction analysis. This chapter focuses on the first iteration, personas elicitation, which proposes a data-driven persona development method (DDPDM) that utilises machine learning, precisely a k-means clustering technique. Data analysis is conducted using two datasets. Eight personas are produced from the two data analyses. The pragmatic findings from this study make two contributions to the current literature. Firstly, the proposed DDPDM uses machine learning, specifically k-means clustering, to build data-driven personas. Secondly, the persona template is designed for university students, which supports the construction of data-driven personas. Future work will cover the second and third iterations.

Game-Based Learning

The continuous growth in the development of interactive technologies has lighted up the game-based learning applications. The game-based learning applications motivate the students to enhance their knowledge and improve the overall student learning experience. Learning with fun and entertainment is the prime aspect of any interactive platform. The skills and knowledge learned by game-based learning are kept longer than traditional learning methods. In addition, an efficient learning method involves students completely in what they are doing. The game-based learning application is very helpful for the physically handicapped students to reveal their intellectual ability. This chapter discusses how the game-based learning applications impacts the Indian education system, national and international status of game-based learning, challenges of game-based learning, existing game-based learning applications, and present and future impacts of game-based learning applications on the Indian education system.

Recommender System Techniques and Approaches to Improve the Modern Learning Challenges

Recommendation systems have been developed from the web. These recommendation systems are useful in collecting information from an available set of sources for a user's preferences. The information can be acquired from user's collection of details to share, to review, to do positive ratings by monitoring the user's behavior to improve the quality of top ‘N' recommendations. Now if we come to modern learning system, it has good framework to influence the training factors from the data, triggers, and learner's preferences. Modern learning can be compared to online learning which carries to the future needs. Modern learning can be instituted in schools, engineering colleges, and working campus. The modern learning system combines interrelated data, processes, and resources to create a system of interdependencies that work together, adapting to changing business needs. These interdependencies include multi-level dynamics driven by the organization, training professionals, technological advances, and the learners themselves.

A Machine Learning Model for Advanced Decision Making in Smart Education Systems

A smart education system uses emerging technologies and generates a vast amount of heterogeneous data in the learning environment. The conventional methods presently used by the educational administrators for decision-making are minimal and take more time to generate the results. The educational administrators could not be able to predict the results quickly and advance for better decision-making. Today, artificial intelligence approaches are widely used in educational systems for automating educational processes. These approaches achieve a better, efficient, and effective modern education system. Integrating machine learning deep learning techniques with a smart education system can automatically analyze the generated data for better decision-making and provide recommendations to students and educational administrators. This chapter aims to introduce a machine learning model to predict the outcomes in a smart education system.

Recommender Systems for E-Learning

The recommender systems are really important in this phase because the users want to be concentrated and to be focused on the domain in which they are interested. There should be minimal deviation in the topics suggested by the recommendation engines. Some of the famous e-learning platforms suggest recommendations based on tags such as highest rated, bestsellers, and so on in various domains. This ultimately makes the users deviate from the domain in which they have to master, and it results in not satisfying the user needs. So, to address this problem, effective recommendation engines will help provide recommendations according to the users by implementing the machine learning techniques such as collaborative filtering and content-based techniques. In this chapter, the authors discuss the recommendation systems, types of recommendation systems, and challenges.

A Comparative Analysis of Blended Models at Tertiary Level

In this research study, the learning outcomes of a blended learning course in an ESL classroom is reported. Although previous research studies have adequately addressed the effect of blended instruction on learning outcomes, there is a dearth of research on the effectiveness of flipped and online model. A review of current literature on online and flipped models revealed that both these models positively impact the learning outcomes. Since this study aims to measure the impact of these two models, an experimental research design was chosen. Two homogeneous groups with a sample size of 22 from each were randomly selected for the study. The instructional method for Group A was the flipped model, and the instructional method for Group B was the online lab model. The t-test results indicated that the the flipped group outperformed the online group.

Personalized Ontology-Based Adaptive E-Learning System

This chapter illustrates novel methods to provide personalized and adaptive content to the learners. This chapter illustrates a new methodology of automatically constructing concept maps using ontology to measure the learners' understanding for a particular topic, thereby teachers can adopt adaptive teaching based on the learners knowledge structures as reflected in the concept maps. The teachers can dynamically revise and deliver instructional materials according to the learners' current progress. In the approach, the authors provide dynamic content to the learners based on neuro fuzzy domain ontology extraction algorithm. This method also provides a personalized ontology model of a learner to learn the ontological user profiles from both world knowledge base and user local instance repositories. The main quality of the innovative work is to mine the personalized ontology of the learners to extract their knowledge through ontology mining using Inc Span+ algorithm.

Learning Management Systems in the Era of E-Learning

The twenty-first century has seen a paradigm shift occasioned by the onset and rapid spread of the coronavirus pandemic in different perspectives. The lockdowns and other protocols introduced by health authorities globally inhibited the progress of many business sectors of the economy. The education sector for example has been affected significantly with e-learning taking a central position to promote social distancing and minimize the spread of the virus. COVID-19 has made the work from home approach a new normal. LMS systems have become effective tools of delivering online content and been considered highly significant on the advent of the pandemic. This chapter presents LMS systems as a medium of e-learning and explains how different digital tools facilitate both synchronous and asynchronous e-learning. A systematic literature review was adopted to shed light on different functionalities of LMS systems and the associated digital tools used for e-learning. The chapter ends by identifying challenges associated with LMS systems and suggesting possible mitigation strategies.