Artificial Intelligence-Enabled Game-Based Learning and Quality of Experience: A Novel and Secure Framework (B-AIQoE)

Abstract

Game-based learning in schools and colleges, with the help of AI-enabled augmented intelligence techniques, is reported to improve children’s neurodevelopment, intellectual sensing, and specific learning abilities, according to US officials. There is currently a huge transformation from traditional assisted learning to augmented reality-enabled computer-based learning. Globally, there has been a dramatic increase in the use of game-based augmented learning in most schools and colleges. A few problems arise that create concern, such as the emerging effects of gaming on institutional premises, the disordering of children’s involvement after game-learning, the rate of learning and attendance, adaptation, and teachers’ experience. To address these individual aspects, we proposed a blockchain Ethereum-enabled, secure AI-based augmented game learning environment, called B-AIQoE, in which protected on-chain and off-chain channels are designed to handle all the transactions and exchanges among students before analysis in terms of color transition, redundancy, unethical transmission, and related vulnerabilities. On the other hand, the proposed system examines and analyzes the Quality of experience (QoE) and improves accessibility as it receives feedback from the students and teachers. For the purpose of automating game-based transactions, three different aspects are discussed, such as verifying and validating student-teacher registration, creating content for game-based learning and privacy, and updating students’ records and exchanges. Finally, this paper separates, analyzes, and discusses a list of emerging challenges and limitations and their possible solutions involved in creating the proposed system.

1. Introduction

Artificial intelligence (AI) has the objective of creating machines that reproduce human capacity in the forms of reasoning, awareness, decision making, and association with problem-solving [1]. The technology stimulates outside and real-time data analysis, which is usually related to the domain of machine learning (ML). AI is sub-divided into two parts, ML and deep learning (DL) [2,3]. ML, on the other hand, is divided into three sub-domains: (i) supervised learning, (ii) unsupervised learning, and (iii) semi-supervised learning. However, the evolution of AI in education creates a new paradigm in terms of modern learning (computer-assisted learning), which increases teachers’ capabilities and allows them to maintain a list of critical tasks that effectively accompany individual students and efficiently support the learning process [4]. Augmented reality (AR) plays a vital role in the development of the educational learning process by providing a holographic-based digital learning environment. Subsequently, the concept of learning by making the change to digital learning through experimentation, which is related to advanced learning experiences, project implementation, hands-on experience of students throughout the learning and creation process, and support of smart technologies [5,6], emerged.

The collaboration of AI with augmented intelligence helps in the creation of an advanced learning environment, specifically personalized teaching. It allows instructors to deliver interactive sessions with the help of a digital-virtual platform, where students access more knowledge as compared to that available on traditional platforms. However, no specific standard applicational model has been proposed which deviates from previous methods of learning and development imposed over decades [7,8]. Based on knowledge delivery and an interactive learning environment, the critical issue that emerges is the standard lifecycle of the learning process, which is not naturally conductive to this new platform. After the emergence of educational intelligence, a method of do-it-yourself education, using several artificial intelligence resources, was developed to improve the teaching environment by enabling experimentation, where students received the teaching benefits at the center of the process of learning [9]. Thus, the evaluation of education 4.0 considered AI-enabled model-based learning (computer-assisted advanced learning, such as virtual reality, etc.), raising various types of challenges and limitations by transforming the experience of learning [10]. In fact, meaningful analysis requires accompanying pedagogical practices and experiences.

The game-based learning environment provides more personalized, flexible, inclusive, and interactive teaching that is enabled through the augmented applications [11]. This advanced learning perspective enhances children’s abilities in terms of critical thinking, generating new ideas, solving real-time challenges, and making future discussions. All of this is made possible by the individualized approach, in which each child sets his or her own pace. The utilization trend is growing day by day and is expected to increase in the future since augmented intelligence makes it possible to amplify artificial intelligence in the educational environment [12], as shown in Figure 1. However, such augmented reality-enabled game-based applications are developed by coding (programming) and allow for the transformation and addition of new content that was previously used in traditional teaching methods, as shown in Figure 1. Because they use different points of view, AI-enabled augmented adaptive platforms have been proposed to strengthen educational systems [13]. This platform’s primary goal is to collect data from students and users related to the searched topic (and record their experiences), analyze individual user experiences, and provide distinct methods of learning based on initial knowledge up to the level of an expert.

To measure the fluctuation of experiences, the system and the student schedule Quality of experience (QoE) evaluations [14]. Thus, it is designed to be integrated into the classroom, offering a range of advantages; for example, it records individual transactions and analyses and serves as new or improved input for instructors to better manage student learning infrastructure. It analyzes the content of teaching, along with the sharing of displays, where all the aspects of data exchange are examined, including the color scheme. However, the increasing number of user interactions increases the load of system analysis and updates the information based on the experience [15]. This improves the student’s skills and abilities to solve critical problems and seeks to complement learning abilities. In a game-based learning environment, for example, it is used to reduce difficult prospects and engage a teacher in identifying and detecting the nearest point of a critical issue for their students. While using AI, ML, augmented intelligence, and DL techniques, complex behavioral patterns related to challenges emerge. As a result, when teachers and students participate in class activities and propose ideas, a large body of derived and shared contents poses serious issues. Thus, a significant procedure is needed to evaluate individual transactions and privacy before delivery of the content.

The role of blockchain-enabling technology in the educational environment has been envisioned and adopted by several developed countries. In this situation, these education sectors are making it possible for kids to learn through games using augmented intelligence in their schools and colleges [16,17,18,19]. Various educational experts are utilizing distributed ledger technology as the modular infrastructure to protect against the exchanging of forged content, unethical data sharing during lectures, and maintaining distorted colors (which likely affect the neurological health of individuals) while students are learning. Such types of issues are usually intended for server-based centralization systems. This technology enables the improvement of the distributed peer-to-peer nodes’ security capabilities by enhancing encryption policies using thresholding hash-based proxy re-encryption. The collaboration and deployment of intrusion detection and secure file storage systems can dynamically install firewalls, tune anti-disclosure procedures, and guarantee the ledger immutability, provenance, and trustworthiness of the augmented learning delivery under the digital educational policies.

This paper discusses the detailed design of the augmented intelligence-enabled process of game-based learning and its related impacts in the educational environment. However, a novel and secure blockchain-based distributed framework, called B-AIQoE, is proposed that enables multimedia tools to receive and dispatch learning content in a protected manner. The design of B-AIQoE provides educational content integrity, transparency, provenance, traceability, and reliability. It ensures proper learning content delivery through augmented devices and other related operations while creating a trusted platform between the teachers and the content manager. The process of distributed game-based learning is classified as follows: (i) content designing, (ii) content implementation, (iii) content deployment, and (iv) content delivery. The proposed B-AIQoE management also protects the privacy of individual node transaction execution events and provides storage with ledger deliverance protocols in the encrypted form of NuCypher-based re-encryption blocks. The main contributions of this paper are discussed as follows:

• In this paper, we propose a distributed blockchain-enabled Ethereum architecture for computer-assisted learning. In this process, the augmented intelligence-based game-based learning system is derived.
• B-AIQoE, an AI and Quality of experience-enabled collaborative approach, is presented, which mainly focuses on the developed learning content analysis and management before deployment and dispatching through the augmented devices.
• A permissionless blockchain network is designed that provides an integrated, secure environment for content delivery. For this purpose, there are two communication channels: on-chain and off-chain.
• Smart contracts are designed, created, and deployed, such as a number of students, along with teachers and content managers (staff), registering, adding content, managing it, updating changes, and exchanging ledgers.
• Finally, we separate and highlight a few emerging issues while creating and deploying B-AIQoE in real-time. In this manner, we provide possible solutions, along with future open research directions.

The rest of this paper is aligned, organized, and structured as follows: In Section 2, various related research is categorized and presented, highlighting the emerging challenges, limitations, and related details. The role of blockchain-enabling technology in learning domains is discussed in Section 3. In Section 4, we present a novel and secure blockchain-enabled distributed framework using Ethereum for computer-assisted learning through augmented intelligence and game-based points of view. However, several implementation issues are listed while deploying B-AIQoE in the real scenario mentioned in Section 5. Finally, we conclude this research paper in Section 6.

2. Related Work

Artificial intelligence-enabled game-based learning creates a new paradigm in the educational environment, especially from the learning perspective, where the younger generation is involved in the utilization of digital technology, such as mobile or ubiquitous devices. The use of augmented technology in early childhood is considered one of the new beginnings of learning activities [18,19]. The adaptation of games in education does not require any additional prerequisite for the younger generation; thus, its use creates a positive impact on the learning scenarios, such as the objective of the system to provide only educational subject matter and not entertainment-related content. In comparison to the traditional method, tutors achieve concrete learning objectives and desired educational goals [18,19]. As shown in

Table 1. Blockchain, AI, augmented intelligence, and game-based learning related literature.

Title of Research	Detail Description	Research Similarities/Dissimilarities	Issues, Challenges, and Limitations
A game-based vocabulary learning for early childhood students [20]	This paper discussed the emergence of a new paradigm in an educational environment, one of game-based learning, especially serious gaming (no entertainment), and its positive effects in the stage of early childhood.	Short-term and long-term vocabulary learning Measure cognitive impacts Rate of learning Teacher-to-student relationship	Content organization and management issues The cost of privacy and security Consume more resources for computation
The role of serious gaming and impacts in character building in early childhood [21]	The authors of this paper emphasize issues in the current design of serious gaming that are used to improve students’ social behavior, character development, and intelligence in the early childhood stage.	Tested on autism spectrum disorder students QoE Less content security	No process hierarchy was designed Centralized system Streamlined automation-related issues
A Collaborative approach: AI with game-based learning [22]	This paper presented the role of AI with game-based learning and personalization for the purposes of detection, recognition, and analysis of human emotion, identifying voice-based records, and intelligence.	Immersive technology in education Use of virtual and augmented reality Constructed with artificial intelligence modules	More computational resource utilization while managing and organizing game-based contents Lack of learning about information privacy and security Cost-inefficient approach
Hidden reflection analysis during game-based learning in school environment [23]	A framework for automating access to students’ written reflections and their responses during the delivery of game-based lectures is proposed.	Self-regulated learning Metacognition Role of natural language processing (NLP) to improve vocabulary Rate of reflection during training	Streamline the delivery of content Quality and maintenance are inefficient Centralized system
Affection detection of game-based learning in schools’ children [24]	This paper investigates the previously published multimodal effects detection framework and separates the main aspects that impact early childhood students while learning through games or serious gaming.	Multimodal prospects Data fusion Affect detection ML-based game-based learning strategy	Sensor-free approach Responsive learning environment Different data channels are designed
AI-enabled collaborative approach is designed for upper elementary classes: game-based learning [25]	The author of this paper proposes a new paradigm by using AI-infused collaborative game-based learning and AI learning experiences in an educational environment, especially in upper elementary classes (K-12).	K-12 AI education Collaborative inquiry strategy	Lack of standardization No process hierarchy of content management and deliverance Scope of data privacy issues

, there is a variety of related work presented that highlights the difference between traditional learning and computer-assisted learning, particularly the use of AI, augmented intelligence, and blockchain distributed ledger technology to improve the educational environment and digital learning prospects.

3. Application of Augmented Intelligence in the Educational Environment

During the COVID-19 pandemic, students and teachers started using AI in the classroom more and more as a way to solve problems and get better at what they do [26]. One of the new problems that many developing countries are facing is how to give everyone the same chance to learn and get an education. Learning, on the other hand, is the fundamental process for developing social and cultural norms, allowing the new generation to construct personalization, skills, knowledge, experiments, and practices that must be passed down to individuals. Thus, education thrives when people work together to build a strong learning community [27].

Currently, the involvement of augmented intelligence in education could improve educators’ professional judgement and learning capabilities because of its proposed innovative designs and features. It saves time for the administration and helps them determine the problems students are having with their sessions and their learning prospects. It calculates the engagement of students by dispatching game-based content in the classroom. Perhaps an examination of which content execution events work to engage students and which do not is in order. The role of game-based learning with augmented intelligence is discussed as follows:

The Impact of Game-Based Learning in Education

The concept of game-based learning is that principles and game characteristics are associated with learning community prospects [28,29,30,31,32,33]. This type of activity promotes student involvement and engagement in the learning domains. However, a few key components, such as badges, quizzes, leader boards, point systems, discussion boards, fastest response platforms, etc., make game-based learning technology more appealing. Teachers can also give students an extra week to complete their final projects or assignments, and the teacher who is in charge of the board at the time takes the lead. These kinds of digital academic rewards also enhance the capability for student learning in the classroom.

Game-based learning is considered an active learning approach in which games are designed to improve the intelligence (IQ level) of the students. This learning technique promotes critical thinking and enhances problem-solving skills. However, the game-based learning technique is categorized into two subparts, such as (i) digital games and (ii) non-digital games, which include experiments and allow students to experience the learning first hand [34,35].

4. Proposed Framework with Working Execution of B-AIQoE

Figure 2 shows the proposed distributed framework, called B-AIQoE. It is an Ethereum blockchain and AI-enabled new and secure infrastructure that was built and put into place with the goal of maintaining the quality of game-based learning content on augmented devices. This collaborative approach to evaluating the user’s experience by collecting feedback is associated with the Quality of experience (QoE) techniques. This proposed B-AIQoE is divided into six different folds. First, the list of augmented devices/wearables is registered along with the registration of stakeholders for the purpose of securing transactions.

Following that, it moves through the game-based learning process hierarchy, which is divided into six parts: content (class-wise data collection) data collection, content design, content implementation, content presentation and dispatch, content updates, and content records (logs) preservation. However, to reduce the consumption of resources, we designed a middle processing unit that provides lightweight authentication using blockchain Ethereum technology and executes events of nodes’ transactions in a resource-limited manner (because of the 4 MB fixed size of the nodes).

However, the designed process hierarchy of game-based learning is connected with the AI-enabled ML technique (SVM) that provides a procedure for quality management, organization, and optimization while preserving the records (content) in the immutable blockchain storage, as shown in Figure 3, whereas the stakeholder registration and read/write facilities are completed and executed through the smart contracts. There are three different contracts that are designed, created, and deployed to automate the execution of on-chain (implicit transactions of the chain) and off-chain (explicit transactions of the chain) related transactions, verification and validation, new content registries, and update content requests. For updating ledger requests, we used Ethereum-enabled proof-of-work and proof-of-work consensus policies with a digital signature, as shown in Figure 4.

The request for content change will be updated after getting 51% consensus permission. However, the logs of adding new records, updates, registrations, etc., are stored in the InterPlanetary File Storage Structure (IPFS), a distributed database that provides storage scalability with a minimal amount of cost as compared to other distributed storage, such as USD 100/month. The other critical components are highlighted as follows (as also mentioned in

Table 2. Comparison between the proposed B-AIQoE with other state of the art frameworks (1).

Categories	Other State of the Art Methods/Models/Frameworks (1) [36]	Other State of the Art Methods/Models/Frameworks (2) [37]	Other State of the Art Methods/Models/Frameworks (3) [38]	The Proposed B-AIQoE
Use of artificial intelligence-enabled machine learning techniques	√	√	√	√
Game-based learning consent	√	√	√	√
Process hierarchy with standardized steps		√		√
Learning content security				√
Augmented reality/devices/intelligence	√		√	√
Blockchain-Web 3.0				√

and

Table 3. Comparison between the proposed B-AIQoE with other state of the art frameworks (2).

Categories	Other State of the Art Methods/Models/Frameworks (1) [36]	Other State of the Art Methods/Models/Frameworks (2) [37]	Other State of the Art Methods/Models/Frameworks (3) [38]	The Proposed B-AIQoE
Encryption/NuCypher threshold re-encryption mechanism for content privacy				√
Quality of experience (QoE)		√	√	√
Integrated/collaborative approach	√	√	√	√

):

• Distributed nodes inter-connectivity
• Smart contracts
• Peer-to-Peer (P2P) network
• Consensus policies
• Ethereum guest fees
• Digital signature
• Cryptographic hash-encryption.

5. Implementation, Simulation, and Discussion

In this context, we mention a few emerging open issues in AI-enabled game-based learning that need to be critically evaluated, and so we formulate such challenging problems and limitations and provide a positive solution as follows:

5.1. Interoperable Platform Limitation

In a peer-to-peer network environment, a public communication channel is designed that handles a number of transactions through either the on-chain or off-chain channels within the chain. However, a number of issues have arisen regarding the cross-chain platform, interoperable chain-to-chain issues, and ledger synchronization-related issues while establishing interconnectivity between different nodes of different chains [39]. This highlighted aspect requires a permanent solution while creating and deploying a blockchain-enabled Ethereum public chain. The proposed B-AIQoE allows node-to-node events of transaction execution within the chain, but needs to maintain communication between different related blockchains for a better learning experience. For analysis of this matter, we designed the interpreted blockchain business process platform; the main assumptions being to secure game-based learning transactions (both read/write), content delivery, educational ledger management, exchange details, and storage through the distributed application (DApp), whereas the current legacy and standard of the proposed serverless environment raise a lack of interoperability. This proposed solution will help to maintain interoperable platform issues and provide teachers and students with interactive communication platforms where information is exchanged between or outside of the chains. However, the blockchain-enabled Ethereum solution of a cross-chain platform manages transparent interconnectivity between nodes during the exchange of learning contents.

5.2. Streamline Automation of Game-Based Contents and Lack of Standardization

Within the game-based learning environment, different types of content are available, such as early childhood, middle school, and adult content. The procedure (including design, implementation, deployment, and dispatch) contributes to the lack of standards while managing game-based learning contents [40,41]. The lifecycle of current computer-assisted learning, from content design to dispatching, is unsecure, repudiated, and less reliable. As a result, it has a negative impact on the procedures of content organization and management, resulting in inconsistencies in quality. Therefore, by robustly standardizing and automating, a collaborative AI-blockchain with a QoE approach is used, which enforces a standard hierarchy and improves content management efficiency and quality.

5.3. Scope of Computer-Assisted Learning and Privacy

There is no proposed standard lifecycle of computer-assisted learning, and so there is no discussion on the real-time management and organization of game-based learning contents [41]. For analysis of such issues, we examined and evaluated different types of AI mechanisms, QoE strategies, and blockchain networks, along with their various versions. We took the best of these and created a B-AIQoE framework-enabled DApp that manages the scalability and preservation of game-based learning contents in real-time. By adopting this method, the proposed B-AIQoE reduces the cost of information privacy. However, there is a need for well-organized management of the learning content hierarchy. Shifting the existing server-based, centrally managed infrastructure to a serverless environment is required for this purpose because the current legacy of scalability, storage, and information exchange is inefficient and incurs additional costs to maintain privacy and security. Allowing a fixed size of node transactions on the distributed network reduces the load on resources and increases security and privacy.

6. Conclusions

The main objective of this paper is to separate and highlight the existing limitations in computer-assisted learning and its related technologies discussed herein. This paper proposes a novel distributed framework, namely, B-AIQoE, a collaborative approach of blockchain and AI-enabled game-based learning processes that mainly aims to deliver educational learning content in a better and more secure manner. However, the designed framework evaluates and measures the quality of experience while learning contents are designed, created, deployed, and dispatched through augmented devices. This complete distributed procedure reduces the resource consumption to a certain extent, as compared to the previous computer-assisted ecosystems (as mentioned in Table 2 and Table 3), including computational power, network bandwidth, and storage, along with the cost of privacy and security. The proposed B-AIQoE has been created to make sure that resources are managed in a safe and cost-effective way, while also maintaining the system’s integrity, transparency, and provenance. In this paper, we have presented three different smart contracts (chain codes) for automating verification and validation, such as the registration of stakeholders (students/teachers), the addition of new learning content, and the process of updating content. Ethereum’s proof-of-work and multi-proof-of-work were created and put to use to cut down on the use of resources, such as computing power, network bandwidth, and the need to keep resources safe. We also created a blockchain-enabled, lightweight authentication mechanism based on Ethereum for this purpose. Individual node transactions events are executed either on-chain or off-chain over the distributed network. However, the new registries, learning contents, and updates are recorded through the designed process of B-AIQoE and preserved in the InterPlanetary File Storage System (IPFS), which is immutable in nature. Finally, this proposed B-AIQoE, a cost-efficient resource consumption system, is considered a good candidate for industrial adaptation.