EduArk

Literature Survey

On account of its accessibility and virtual aspect, online education has grown substantially in popularity in recent years. The lack of customized connection between teachers and pupils, however, is one of the issues that online education systems must deal with. This literature study aims to investigate the idea behind a smart, personalized education platform driven by AI created to deal with this problem. The review has been broken down into four sections, each of which focuses on different areas of features and goals[1][2].

[1] Hellas, A.; Ihantola, P.; Petersen,A.; Ajanovski, V.V.;Gutica, M.; Hynninen, T.; Knutas, A.; Leinonen, J.; Messom, C.; Liao, S.N. Predicting academic performance: A systematic literature review. In Proceedings of the Companion of the 23rd Annual ACM Conference on Innovation and Technology in Computer Science Education, Larnaca, Cyprus, 2–4 July 2018; pp. 175–199.
[2] T. Huang, Y. Mei, H. Zhang, S. Liu, and H. Yang, "Fine-grained Engagement Recognition in Online Learning Environment," in 2019 IEEE 9th International Conference on Electronics Information and Emergency Communication (ICEIEC), Jul. 2019, pp. 338-341. doi: 10.1109/ICEIEC.2019.8784559.

Research Gap

Current research lacks a comprehensive exploration of seamlessly integrating expected marks prediction models within a multi-tenant educational platform. Achieving a balance between personalized predictions and ensuring data privacy across diverse institutions remains an unexplored area. Additionally, the research landscape falls short in providing a nuanced understanding of how to effectively cater to individual learning styles and preferences when formulating personalized learning plans.
Incorporating real-time feedback mechanisms for continual refinement is also a crucial yet underexplored aspect. Moreover, limited studies have delved into the development of dynamic algorithms that can adaptively recommend the most suitable assessment approaches based on contextual factors such as subject matter, student performance, and learning objectives. There is also a dearth of research on seamlessly incorporating technology-enhanced assessments into the recommendation process. Furthermore, the existing research overlooks the need for comprehensive student profiles that encompass academic performance, interests, extracurricular activities, and career aspirations. Aligning subject stream recommendations with a student's envisioned long-term career trajectory is an unexplored territory that holds potential for significant impact.

Research Problem & Solution

01. How does a multi-tenant architecture impact data security and privacy in E-Learning systems?

02. What are the key factors influencing a student's performance in a given semester/term?

03. What learning strategies are most effective for different types of students (e.g., visual learners, auditory learners, etc.)?

04. What factors influence the effectiveness of different assessment approaches in evaluating student learning outcomes?

05. What is the correlation between subject stream selection and long-term career success?

Research Objectives

Multi-Tenant

To analyze the level of customization and personalization options that can be provided to individual institutions within a shared multi-tenant infrastructure.

Student Performance

To develop and validate a machine learning model for predicting students' expected marks for the upcoming semester/term based on historical performance data and relevant contextual factors.

Learning Plan

Building up a module to identify suited learning strategy out of Video, Audio, Text for the Student.

Assessment Approach

Research and evaluate various assessment methods, including exams to determine their effectiveness in evaluating student learning outcomes across different subjects and contexts.

Subject Stream

provide a clear roadmap for developing systems that can effectively recommend learning plans, assessment approaches, and subject streams for students based on their individual profiles and educational objectives.

Research Methodology

The system in question is a robust multi-tenant platform designed to accommodate a diverse user base, comprising Super Admins, Admins, Teachers, and Students. This comprehensive system features two distinct client applications: the Backoffice Angular SPA, accessible exclusively to Super Admins, and the “Eduark” Application, tailored for use by Admins, Teachers, and Students. The former serves as a powerful administrative hub, enabling system-wide configurations, user management, and comprehensive monitoring. Meanwhile, Admins wield significant authority within their respective tenant domains. They can proficiently manage users, classes, curriculum, and tenant-specific settings through the “Eduark” Application. Additionally, Admins have access to the Master API, dedicated to handling administrative functions. Teachers leverage the “Eduark” Application to seamlessly facilitate educational activities, including creating and managing classes, assigning tasks, monitoring student progress, and conducting interactive sessions. Students primarily interact with the “Eduark” Application for class participation, assignment completion, and personal progress monitoring, with access limited to their individual data and class materials. The system's architecture adheres to the principles of clean architecture, meticulously segregating concerns across distinct layers for optimal modularity, testability, and scalability. This encompasses the Presentation layer for user interfaces, Application for business logic, Domain for core business entities and rules, and Infrastructure for external interactions. It also features two essential Web APIs: the “Eduark” API, dedicated to servicing “Eduark” Application functionalities, and the Master API, managing system-wide administrative functions, accessible to Super Admins and potentially other authorized users. In addition to these components, the system incorporates a Student Mark Analysis Machine Learning API, augmenting the platform with advanced analytical capabilities. Azure Blob Storage is employed for efficient management of unstructured data, including media files and documents. The database design is structured around a dual-tiered approach, consisting of a Master Database housing shared data and configurations universal to all tenants, and individual Tenant Databases, each dedicated to tenant-specific data. This includes class rosters, assignments, student records, and other customized information tailored to the unique requirements of each client. This comprehensive system architecture ensures a seamless educational experience, offering a robust suite of tools and functionalities tailored to the needs of each user group and tenant.