This paper explores the design and implementation of AI-driven interactive concept maps as components of intelligent textbooks and digital learning environments.

Made with Rinto — analyse your own content free

AI-powered Concept Maps Need

Traditional concept maps lack interactivity and scalability, necessitating enhancements with AI capabilities and interactive features in digital learning environments.

Visualization Role in Learning

Numerous studies have highlighted the pivotal role of visualization in enhancing the comprehension and retention of educational content.

Early Visualization Studies

Early work by Naps et al. emphasized visual engagement, particularly in computer science education, contributes significantly to student understanding and motivation.

Navigable Concept Map Systems

The CoMPASS project and ActiveMath platform introduced navigable and adaptive visual structures to improve structural understanding and navigation efficiency.

Interactive Features Standard

Hollingsworth and Narayanan argue that interactive features like concept maps should be standard components of digital textbooks.

Self-Regulated Learning Support

Barria-Pineda et al. developed visualization tools supporting self-regulated learning via concept-level mapping.

Ontology-Oriented Learning Systems

In ontology-oriented learning systems, concept maps serve as structured, navigable knowledge representations, exemplified by TM4L.

Concept Maps Improve Learning

Many studies have shown that concept maps improve learning outcomes, particularly for complex material.

Meta-analysis Effectiveness

A meta-analysis by Schroeder et al. reviewed 142 studies and found concept maps significantly more effective than reading texts (g = .39) or reviewing outlines (g = .28).

Anatomy Course Study

Bolatli and Bolatli reported higher post-test scores and lower cognitive load in anatomy students using predefined concept maps.

Interactive Glossary Engagement

Elgendi and Shaffer demonstrated interactive glossary maps increased student engagement and repeated glossary use in a computer science e-textbook.

Booc.io System Features

Schwab et al. introduced booc.io, a system with drill-down hierarchical concept maps supporting adaptive navigation and targeted feedback.

LLM Concept Map Construction

Ma and Chen proposed an LLM framework for automated concept map construction from e-books, including segmentation, key concept extraction, and relationship identification.

Causal Concept Maps

Kluga et al. integrated causal concept maps into an intelligent anatomy textbook for personalized navigation, quiz adaptation, and content feedback.

Legal Concept Hierarchy System

Wehnert et al. presented a dynamic visualization system for exploring concept hierarchies extracted from legal textbooks, supporting various navigation modes.

Interactive Hierarchical Concept Map Platform

This project explores the implementation and evaluation of an interactive, hierarchical concept map platform designed to enhance digital learning experiences.

Interface Layered LLM Design

The interface introduces a layered, LLM-augmented design supporting on-demand concept generation, infinite drill-down navigation, and in-context information display.

Current System vs Ma and Chen

Our system, similar to Ma and Chen, uses LLMs for concept map generation but addresses interactive UI/UX design and cognitive overload.

Aspect	Ma and Chen System	Proposed System
LLM Use	Automates extraction of concepts and relationships	Employs LLMs to generate concept maps
Real Educational Context	Does not address use in real educational contexts	Addresses use with validation based on real student feedback
Interactive UI/UX Design	Lacks production-ready interactive UI/UX design	Focuses on production-ready interactive UI/UX design
Cognitive Overload	Overlooks cognitive overload from large, flat maps	Addresses via hierarchical drill-down navigation and embedded pedagogical content
Content Alignment	Uses full course materials	Plans to incorporate RAG in future iterations

System Applied Courses

The system was applied in Object-Oriented Programming and Data Structures, and UI Design and AI-Assisted Frontend Development university courses.

Main Course Map Overview

The Main Map of the Course functioned as a central, interactive overview of the entire curriculum, visually organizing key topics and their relationships.

Information Panel Details

The Information Panel provided contextual details for each concept selected within the map, displaying concise explanations, examples, or supporting materials.

Child Concept Map Expansion

A Child Concept Map is a secondary, more focused map that expands upon a specific node from the main concept map.

Infinite Drill-Down AI Exploration

The Infinite Drill-Down AI-Based Domain Exploration feature enables learners to move beyond predefined content by dynamically generating new sub-maps using AI.

AI-Driven Map Generation Process

The system generates interactive concept maps by querying LLMs, such as GPT-4o-mini, based on provided course metadata and prompt-tuning commands.

Concept Map Development Steps

The end-to-end concept map development process includes eight distinct steps, from course creation and initial generation to refinement and nested map creation.

Quality and Coherence Assurance

The refinement process requires manual review of each map level and concept description to validate pedagogical relevance and accuracy.

Refinement Process Time

For highly customized courses, the refinement process can be time-consuming; however, AI-assisted generation significantly reduces the baseline effort.

Limitations of Generic LLM Maps

Several limitations emerge from relying solely on large language models (LLMs) without incorporating course-specific materials.

AI Concept Map System Architecture

The system features a lightweight, modular architecture that combines AI-driven content generation with interactive frontend visualization.

Python Backend

The Python backend handles HTTP requests, interacts with the OpenAI GPT-4o-mini API, and stores course content in a file-based structure.

Vis.js Frontend

The frontend uses Vis.js to render dynamic, clickable concept maps, enabling content display and drill-down navigation.

Prompt-Engineered Map Generation

Concept maps are generated by dynamically constructing and sending prompts to an LLM, such as GPT-4o-mini, driven by frontend input and course metadata.

Prompt-Engineered Concept Pages

The system generates individual concept pages as HTML fragments through targeted prompt engineering when a user selects a concept node.

Content Management Panel

The Content Management Panel is a core instructor-facing tool enabling efficient oversight of all AI-generated concept map content.

Evaluation: Student Feedback

This section presents the design and results of a student survey evaluating the perceived effectiveness of AI-powered concept map interfaces.

Survey Design Overview

Hierarchical concept maps were developed for two courses at American University Kyiv, and a survey was administered to students.

Survey Results Summary

The analysis shows high student satisfaction and perceived effectiveness, quantitatively confirming that the interactive and visual structure of concept maps was well-received.

Overall Satisfaction Score

Students reported an average satisfaction score of 8.91 out of 10 with their learning experience using the concept map app.

Learning Effectiveness Score

Students rated the app’s usefulness for learning new content at an average of 8.31 out of 10.

Engagement & Enjoyment

A large majority (91.4%) of students agreed that the app made learning more engaging or enjoyable.

Drill-Down Usefulness

100% of students confirmed that the ability to click on concept names and navigate through nested maps was helpful.

Preferred Review Methods

Students expressed varying preferences for reviewing course material using interactive concept maps versus traditional materials.

Most Useful Features

Open-ended responses revealed clickability/depth navigation and visual structure/relationships between topics as major advantages.

Suggested Improvements

Students suggested improvements related to UI, current interactions, and additional functionalities.

Evaluation Conclusion

High ratings across both courses suggest AI-generated maps are relevant for learning, and the hierarchical drill-down structure offers an effective human-computer interaction pattern.

Conclusion and Future Work

This study presented the design and implementation of an AI-driven system for generating interactive hierarchical concept maps for digital learning environments and intelligent textbooks.

System as Intelligent Textbook

The system positions itself as a practical implementation of the intelligent textbook paradigm, augmenting content delivery and navigational structure for student-centered learning.

Evaluation Results Summary

Evaluation results, based on 69 AI-generated concept maps and 245 unique concept descriptions across two undergraduate computer science courses, confirm the system's educational value.

Key Findings Reinforcement

Feedback from 35 student surveys underscored the potential of integrating AI-generated maps into intelligent textbooks and online courses to enhance comprehension, engagement, and structured review.

Future Work Directions

Future work will address current limitations identified in the system’s architecture and usage, focusing on several key directions.

Acknowledgements

The author thanks American University Kyiv instructors Roman Tymoshuk, Andrii Tsabanov, and Ivan Danilov for reviewing and integrating the concept maps into the learning process.

Generative AI Declaration

The author used ChatGPT-4 and Grammarly to check grammar, spelling, improve writing style, paraphrase, and reword, taking full responsibility for the content.

▸ 10 Expand

APEX

AI-driven Interactive Hierarchical Concept Maps

This paper explores the design and implementation of AI-driven interactive concept maps as components of intelligent textbooks and digital learning environments.

Made with Rinto — analyse your own content free

▸ 1 Expand

CONC

AI-powered Concept Maps Need

Traditional concept maps lack interactivity and scalability, necessitating enhancements with AI capabilities and interactive features in digital learning environments.

▸ 1 Expand

CONC

Proposed AI Concept Map System

The proposed system leverages artificial intelligence to generate initial concept maps, refined through human-in-the-loop input for pedagogical accuracy and relevance.

DETL

Interactive System Features

A custom-designed interface enables learners to explore concepts at varying levels of depth, promoting active engagement and quick navigation.

▸ 6 Expand

CONC

Visualization Role in Learning

Numerous studies have highlighted the pivotal role of visualization in enhancing the comprehension and retention of educational content.

DETL

Early Visualization Studies

Early work by Naps et al. emphasized visual engagement, particularly in computer science education, contributes significantly to student understanding and motivation.

DETL

Navigable Concept Map Systems

The CoMPASS project and ActiveMath platform introduced navigable and adaptive visual structures to improve structural understanding and navigation efficiency.

JUST

Interactive Features Standard

Hollingsworth and Narayanan argue that interactive features like concept maps should be standard components of digital textbooks.

DETL

Self-Regulated Learning Support

Barria-Pineda et al. developed visualization tools supporting self-regulated learning via concept-level mapping.

DETL

Ontology-Oriented Learning Systems

In ontology-oriented learning systems, concept maps serve as structured, navigable knowledge representations, exemplified by TM4L.

▸ 7 Expand

FIND

Concept Maps Improve Learning

Many studies have shown that concept maps improve learning outcomes, particularly for complex material.

STAT

Meta-analysis Effectiveness

A meta-analysis by Schroeder et al. reviewed 142 studies and found concept maps significantly more effective than reading texts (g = .39) or reviewing outlines (g = .28).

STAT

Anatomy Course Study

Bolatli and Bolatli reported higher post-test scores and lower cognitive load in anatomy students using predefined concept maps.

DETL

Interactive Glossary Engagement

Elgendi and Shaffer demonstrated interactive glossary maps increased student engagement and repeated glossary use in a computer science e-textbook.

DETL

Booc.io System Features

Schwab et al. introduced booc.io, a system with drill-down hierarchical concept maps supporting adaptive navigation and targeted feedback.

▸ 2 Expand

DETL

LLM Concept Map Construction

Ma and Chen proposed an LLM framework for automated concept map construction from e-books, including segmentation, key concept extraction, and relationship identification.

STAT

GPT-4o Performance

Their evaluation of GPT-4o demonstrated strong performance, effectively extracting key concepts and accurately identifying hierarchical and cross-topic connections.

INSG

LLM Map Structure Insight

LLMs could generate concept maps that differ from textbook structure, reflecting a more logical and content-based organization.

DETL

Causal Concept Maps

Kluga et al. integrated causal concept maps into an intelligent anatomy textbook for personalized navigation, quiz adaptation, and content feedback.

DETL

Legal Concept Hierarchy System

Wehnert et al. presented a dynamic visualization system for exploring concept hierarchies extracted from legal textbooks, supporting various navigation modes.

▸ 7 Expand

CONC

Interactive Hierarchical Concept Map Platform

This project explores the implementation and evaluation of an interactive, hierarchical concept map platform designed to enhance digital learning experiences.

INSG

Interface Layered LLM Design

The interface introduces a layered, LLM-augmented design supporting on-demand concept generation, infinite drill-down navigation, and in-context information display.

CMPR

Current System vs Ma and Chen

Our system, similar to Ma and Chen, uses LLMs for concept map generation but addresses interactive UI/UX design and cognitive overload.

Aspect	Ma and Chen System	Proposed System
LLM Use	Automates extraction of concepts and relationships	Employs LLMs to generate concept maps
Real Educational Context	Does not address use in real educational contexts	Addresses use with validation based on real student feedback
Interactive UI/UX Design	Lacks production-ready interactive UI/UX design	Focuses on production-ready interactive UI/UX design
Cognitive Overload	Overlooks cognitive overload from large, flat maps	Addresses via hierarchical drill-down navigation and embedded pedagogical content
Content Alignment	Uses full course materials	Plans to incorporate RAG in future iterations

EXMP

System Applied Courses

The system was applied in Object-Oriented Programming and Data Structures, and UI Design and AI-Assisted Frontend Development university courses.

▸ 1 Expand

SUBC

Main Course Map Overview

The Main Map of the Course functioned as a central, interactive overview of the entire curriculum, visually organizing key topics and their relationships.

DETL

Main Map Functionality

Each concept node could be expanded to explore subtopics, allowing students to see the “big picture” and access detailed content.

▸ 1 Expand

SUBC

Information Panel Details

The Information Panel provided contextual details for each concept selected within the map, displaying concise explanations, examples, or supporting materials.

INSG

Information Panel Benefit

This feature allowed learners to engage with the content without leaving the map interface, maintaining flow and minimizing distractions.

▸ 1 Expand

SUBC

Child Concept Map Expansion

A Child Concept Map is a secondary, more focused map that expands upon a specific node from the main concept map.

JUST

Child Map Purpose

Child maps support hierarchical learning by allowing users to progressively explore concepts at increasing levels of depth without overwhelming them.

▸ 2 Expand

SUBC

Infinite Drill-Down AI Exploration

The Infinite Drill-Down AI-Based Domain Exploration feature enables learners to move beyond predefined content by dynamically generating new sub-maps using AI.

DETL

Dynamic Sub-map Generation

When a student reaches a terminal node and seeks further explanation, the system can generate an extended concept map based on the topic's semantic description.

INSG

Adaptive Learning Environment

This feature transforms the concept map into an adaptive learning environment capable of supporting personalized, self-directed exploration across limitless depth within a subject domain.

▸ 3 Expand

CONC

AI-Driven Map Generation Process

The system generates interactive concept maps by querying LLMs, such as GPT-4o-mini, based on provided course metadata and prompt-tuning commands.

▸ 8 Expand

DETL

Concept Map Development Steps

The end-to-end concept map development process includes eight distinct steps, from course creation and initial generation to refinement and nested map creation.

DETL

Step 1: Course creation

The instructor defines a new course by providing a course title, a free-form course description, and optional prompt-tuning instructions.

DETL

Step 2: Initial map generation

The system queries the LLM to produce a draft map of key concepts and their relationships.

DETL

Step 3: Full map regeneration

If the initial map is unsatisfactory, instructors can clear it and regenerate a new one from scratch.

DETL

Step 4: Map refinement

Users can ask a secondary LLM (e.g., ChatGPT) to rephrase or restructure parts of the map for better clarity or alignment.

DETL

Step 5: Concept information generation

Clicking a concept opens an interactive panel that fetches descriptive content via the LLM.

DETL

Step 6: Concept information regeneration

Instructors can discard and regenerate concept information if needed.

DETL

Step 7: Concept information refinement

Generated descriptions can be edited or enhanced using a separate LLM interface (e.g., ChatGPT).

DETL

Step 8: Nested concept map generation

Instructors can expand individual concepts by generating subordinate maps, enabling drill-down exploration.

INSG

Quality and Coherence Assurance

The refinement process requires manual review of each map level and concept description to validate pedagogical relevance and accuracy.

JUST

Refinement Process Time

For highly customized courses, the refinement process can be time-consuming; however, AI-assisted generation significantly reduces the baseline effort.

▸ 3 Expand

CONC

Limitations of Generic LLM Maps

Several limitations emerge from relying solely on large language models (LLMs) without incorporating course-specific materials.

▸ 1 Expand

SUBC

Metadata-Based Generation Risks

Initial concept maps are generated based only on a course title and concise metadata, risking shallow or misaligned content in specialized domains.

DETL

Future Work RAG

Future work will explore integrating retrieval-augmented generation (RAG) techniques to enable concept map generation based directly on actual course content.

▸ 3 Expand

SUBC

Infinite Drill-Down Challenges

Although infinite drill-down capability is innovative, it also poses challenges like disorientation and topic drift.

DETL

Disorientation Risk

As users delve deeper into nested submaps, they may lose awareness of their location within the overall concept structure.

DETL

Topic Drift Risk

In the absence of clear semantic boundaries, LLMs may generate tangential or unrelated subtopics, resulting in conceptual divergence.

DETL

Future UX Strategies

Future work will explore introducing depth limits and developing UX strategies for clearer visual cues to help users maintain orientation.

▸ 1 Expand

SUBC

Submap Consistency Issues

Because submaps are generated independently, a concept node on a higher-level map may not reproduce its child nodes when opened in a dedicated submap.

DETL

Algorithmic Consistency Support

Scalable resolution for submap consistency will require algorithmic support and improved tooling across the concept hierarchy.

▸ 5 Expand

CONC

AI Concept Map System Architecture

The system features a lightweight, modular architecture that combines AI-driven content generation with interactive frontend visualization.

DETL

Python Backend

The Python backend handles HTTP requests, interacts with the OpenAI GPT-4o-mini API, and stores course content in a file-based structure.

DETL

Vis.js Frontend

The frontend uses Vis.js to render dynamic, clickable concept maps, enabling content display and drill-down navigation.

▸ 3 Expand

SUBC

Prompt-Engineered Map Generation

Concept maps are generated by dynamically constructing and sending prompts to an LLM, such as GPT-4o-mini, driven by frontend input and course metadata.

DETL

LLM Prompt Instructions

The prompt instructs the LLM to generate at least 15 nodes, define a single root concept, and create a tree-like hierarchical structure without cycles or dangling nodes.

DETL

Detailed Prompt Requirements

The prompt also requires formatting output as Vis.js code, labeling all edges with clear semantic relationships, ensuring domain-specific accuracy, and including a concrete output example.

DETL

Map Caching

The LLM response is cached as a plain text file, serving subsequent requests directly from the local file system to avoid redundant OpenAI API calls.

▸ 1 Expand

SUBC

Prompt-Engineered Concept Pages

The system generates individual concept pages as HTML fragments through targeted prompt engineering when a user selects a concept node.

DETL

HTML Generation Requirements

The prompt instructs the model to generate valid HTML fragments using semantic tags, format code snippets with pre/code, and mathematical content with LaTeX/MathJax.

▸ 1 Expand

SUBC

Content Management Panel

The Content Management Panel is a core instructor-facing tool enabling efficient oversight of all AI-generated concept map content.

DETL

Instructor Management Capabilities

Instructors can create/manage courses, regenerate/refine maps, and customize individual concept pages using third-party LLMs for better alignment.

▸ 2 Expand

CONC

Evaluation: Student Feedback

This section presents the design and results of a student survey evaluating the perceived effectiveness of AI-powered concept map interfaces.

▸ 5 Expand

SUBC

Survey Design Overview

Hierarchical concept maps were developed for two courses at American University Kyiv, and a survey was administered to students.

▸ 2 Expand

EXMP

Course 1: OOP & Data Structures

Concept maps were developed for Object-Oriented Programming and Data Structures for first-year Bachelor of Software Engineering/AI and Data Science students.

STAT

Course 1 Content Count

42 concept maps and 117 individual concept pages were generated and integrated into Course 1.

STAT

Course 1 Survey Participants

24 students participated in the survey for Course 1.

▸ 2 Expand

EXMP

Course 2: UI Design & AI Frontend

Concept maps were developed for UI Design and AI-Assisted Frontend Development for first-year Bachelor of Software Engineering and AI program students.

STAT

Course 2 Content Count

27 concept maps and 128 concept pages were generated and refined for Course 2.

STAT

Course 2 Survey Participants

11 students participated in the survey for Course 2.

STAT

Total Survey Responses

A total of 35 surveys were received from both courses.

DETL

Survey Administration

The survey was proposed at the final part of the course as a means of recalling course content and evaluating the concept map’s effectiveness for structured review.

DETL

Key Survey Questions

Questions assessed overall satisfaction, engagement, effectiveness for review/learning, and usefulness of drill-down navigation.

▸ 8 Expand

SUBC

Survey Results Summary

The analysis shows high student satisfaction and perceived effectiveness, quantitatively confirming that the interactive and visual structure of concept maps was well-received.

STAT

Overall Satisfaction Score

Students reported an average satisfaction score of 8.91 out of 10 with their learning experience using the concept map app.

STAT

Learning Effectiveness Score

Students rated the app’s usefulness for learning new content at an average of 8.31 out of 10.

STAT

Engagement & Enjoyment

A large majority (91.4%) of students agreed that the app made learning more engaging or enjoyable.

STAT

Drill-Down Usefulness

100% of students confirmed that the ability to click on concept names and navigate through nested maps was helpful.

▸ 3 Expand

DETL

Preferred Review Methods

Students expressed varying preferences for reviewing course material using interactive concept maps versus traditional materials.

STAT

Both Equally Effective

60% of students found both interactive concept maps and traditional materials equally effective for review and recall.

STAT

Preferred Interactive Maps

34.3% of students preferred interactive concept maps for reviewing course material.

STAT

Preferred Text-Based Materials

Only 5.7% of students preferred text-based materials alone for reviewing course material.

▸ 2 Expand

DETL

Most Useful Features

Open-ended responses revealed clickability/depth navigation and visual structure/relationships between topics as major advantages.

DETL

Clickability and Depth Navigation

11 students mentioned appreciating the ability to click, navigate, and explore deeper levels of topics.

DETL

Visual Structure and Relationships

9 students highlighted the clarity and usefulness of the visual concept map structure and topic relationships.

▸ 4 Expand

DETL

Suggested Improvements

Students suggested improvements related to UI, current interactions, and additional functionalities.

STAT

App Perfect or Good

7 students felt the app was already perfect or good enough, stating “everything is perfect” or “all is good”.

STAT

UI Improvement Suggestions

8 students recommended UI improvements, including design changes for a more pleasant visual side and a dark theme.

STAT

Interaction Improvement Suggestions

6 students recommended improvements in current interactions, citing issues with dragging, zooming, and scrolling the graph.

DETL

Additional Functionality Ideas

Various individual ideas were suggested, such as online collaboration, shared comments, quizzes, and the ability to ask questions.

INSG

Evaluation Conclusion

High ratings across both courses suggest AI-generated maps are relevant for learning, and the hierarchical drill-down structure offers an effective human-computer interaction pattern.

▸ 4 Expand

CONC

Conclusion and Future Work

This study presented the design and implementation of an AI-driven system for generating interactive hierarchical concept maps for digital learning environments and intelligent textbooks.

INSG

System as Intelligent Textbook

The system positions itself as a practical implementation of the intelligent textbook paradigm, augmenting content delivery and navigational structure for student-centered learning.

DETL

Evaluation Results Summary

Evaluation results, based on 69 AI-generated concept maps and 245 unique concept descriptions across two undergraduate computer science courses, confirm the system's educational value.

INSG

Key Findings Reinforcement

Feedback from 35 student surveys underscored the potential of integrating AI-generated maps into intelligent textbooks and online courses to enhance comprehension, engagement, and structured review.

▸ 7 Expand

SUBC

Future Work Directions

Future work will address current limitations identified in the system’s architecture and usage, focusing on several key directions.

DETL

RAG Integration for Content

One key direction is the integration of retrieval-augmented generation (RAG) techniques to allow the LLM to incorporate course-specific materials into the generation process.

DETL

Usability Enhancements

To improve usability and mitigate disorientation, future enhancements will include research on depth-limiting mechanisms and UX improvements for visual context and navigational cues.

DETL

Submap Consistency Resolution

Algorithmic methods and instructor-facing tools will be developed to support structural validation and map coherence across different levels of hierarchy.

DETL

Refinement Process Integration

The refinement process will be more tightly integrated into the instructor-facing UI for seamless editing and regeneration of maps and concept pages within the same workflow.

DETL

Admin Dashboard Extension

The admin dashboard will be extended to provide broader control and monitoring of content quality.

DETL

Other Domain Exploration

Additional research will explore the system’s effectiveness in other domains such as mathematics, management, and postgraduate software engineering education.

DETL

Agentic AI Workflows

The implementation of agentic AI workflows, where multiple autonomous agents coordinate multi-step concept map construction and refinement, will be explored.

DETL

Acknowledgements

The author thanks American University Kyiv instructors Roman Tymoshuk, Andrii Tsabanov, and Ivan Danilov for reviewing and integrating the concept maps into the learning process.

DETL

Generative AI Declaration

The author used ChatGPT-4 and Grammarly to check grammar, spelling, improve writing style, paraphrase, and reword, taking full responsibility for the content.