Project Type: Game-Based Learning | AI and Machine-Learning Literacy | Interactive Educational Technology
Project Status: Functional Browser-Based Prototype — Developer Play-Testing Completed
https://adewole-data-detective.edtechcreator.chatgpt.site/
1. Project Title
Who Wants to Be a Data Detective? Game-Based Learning for AI and Machine-Learning Literacy
Who Wants to Be a Data Detective? is an interactive educational game I developed to support the understanding and application of foundational artificial intelligence (AI) and machine-learning (ML) concepts.
Inspired by the familiar multiple-choice game-show format of Who Wants to Be a Millionaire?, the application transforms AI/ML coursework into a scenario-based learning experience featuring progressive questions, lifelines, explanatory feedback, virtual points, and audio effects.
The project demonstrates how game-based learning and AI-assisted software development can be combined to create an engaging instructional resource for learners with different levels of technical experience.
During my first graduate residency, I observed considerable differences in participants’ prior knowledge and technical experience. Some classmates had substantial backgrounds in computing and related fields, while others, including me, were encountering many foundational AI and machine-learning concepts for the first time.
This experience highlighted an instructional challenge: learners entering AI-related programs from different academic and professional backgrounds may require additional opportunities to revisit foundational concepts, practice applying new knowledge, and identify gaps in their understanding.
Concepts such as data cleaning, representation bias, class imbalance, data leakage, algorithm selection, and model evaluation require more than memorizing definitions. Learners must also recognize when these concepts apply and make appropriate decisions in realistic situations.
I initially developed practice questions from my coursework to support my own learning. However, I recognized that these materials could be transformed into a reusable educational resource for classmates, newcomers to AI/ML, and individuals preparing for technical interviews.
The central design challenge was to transform conventional AI/ML review questions into an interactive, supportive learning experience that encourages retrieval practice, conceptual reasoning, and application without creating unnecessary time pressure.
This motivated the development of Who Wants to Be a Data Detective?
Who Wants to Be a Data Detective? is a browser-based educational game that uses a progressive question ladder to support the review and application of foundational AI/ML concepts.
The game contains a 50-question bank organized into five categories, with ten questions per category:
Education scenarios.
Industry scenarios.
Definitions and foundational concepts.
Method selection.
Interpretation of results.
Each game selects 15 questions across three editorially assigned difficulty bands:
Foundations: Questions emphasizing terminology and essential concepts.
Application: Questions requiring learners to apply concepts to specific situations.
Decision-Making: Questions involving evaluation, interpretation, and selection among possible approaches.
These difficulty categories are instructional classifications and have not been empirically validated.
Core Game Features
15-Level Progression Ladder
Learners advance through a sequence of 15 questions, earning virtual points for correct responses.
The game includes safety milestones at 1,000 and 32,000 points, with a maximum of 1,000,000 virtual points.
These points are fictional game rewards and have no monetary value.
Four Single-Use Lifelines
Learners can access four optional forms of support:
50:50: Reduces the available answer options.
Ask the Audience: Displays simulated audience responses that may be incorrect.
Call a Coach: Provides a written hint to support reasoning.
Switch Question: Replaces the current question.
Each lifeline can be used once per game.
Explanatory Feedback
After answering, learners can review explanations for all four response options.
The explanations clarify why an option is appropriate or inappropriate, supporting reflection beyond simply identifying the correct answer.
Self-Paced Learning
The game has no countdown timer. Learners can consider their responses, use available assistance, and review explanations at their own pace.
Audio and Visual Engagement
The application incorporates sound effects, optional background music, and volume controls to recreate selected elements of the game-show experience.
Progress Saving and Learning Records
Where supported by the browser, the application preserves game progress locally. Learners can also download a learning record for personal review.
The game does not train or deploy a machine-learning model. Its purpose is to support AI/ML literacy through interactive practice.
The overarching objective was to design and develop an engaging, accessible learning resource that helps learners review and apply foundational AI/ML concepts.
The specific objectives were to:
Transform my AI/ML coursework notes, quiz materials, and scenario responses into a reusable educational resource.
Support the recall and explanation of foundational AI/ML terminology.
Provide realistic scenarios requiring learners to identify data problems and select appropriate methods.
Develop formative assessment questions that require interpretation, reasoning, and decision-making.
Provide explanatory feedback for correct and incorrect response options.
Incorporate optional instructional support through lifelines and hints.
Use game-based motivational elements to encourage continued participation.
Support self-paced learning without time restrictions.
Enable learners to review their performance and identify areas requiring additional study.
Apply AI-assisted development tools to create, test, refine, and publish a functional educational game.
These objectives reflect the intended instructional functions of the resource. Formal evidence of improved learning outcomes or interview readiness has not yet been established.
The development process involved audience analysis, instructional content development, interactive prototyping, game-based design, functional testing, and iterative refinement.
5.1 Audience Analysis and Identification of Learning Needs
The project originated from observations made during my graduate residency.
Participants entered the learning environment with different levels of familiarity with AI, machine learning, and data-related concepts.
I identified a need for a resource that could support learners who required additional opportunities to practice foundational concepts while also accommodating individuals seeking more advanced application and review.
The intended audience includes classmates from different academic backgrounds, newcomers to AI/ML, and individuals reviewing foundational concepts for interviews.
These observations informed the design. They were not collected through a formal research needs assessment.
5.2 Instructional Content Development
I began with my coursework notes, quiz materials, and original responses to the Data Challenge assignment.
The initial content addressed topics such as data quality, cleaning, bias, class imbalance, privacy, and model evaluation.
I used ChatGPT to assist with question development, explanation writing, content organization, and revision.
I reviewed the questions and explanations against my course materials and instructional intentions.
The content was subsequently expanded into a 50-question bank organized around five themes.
Questions include scenarios drawn from education and industry contexts, such as hiring, retail, manufacturing, and fraud detection.
These scenarios were designed to encourage learners to connect technical concepts with practical decisions.
5.3 Instructional Design and Game-Based Learning
The instructional design combines retrieval practice, scenario-based application, formative feedback, and optional learning support.
Retrieval Practice
Multiple-choice questions provide opportunities for learners to recall and apply previously studied information.
Scenario-Based Learning
Realistic situations require learners to interpret problems and identify suitable approaches rather than rely exclusively on definitions.
Explanatory Feedback
Feedback addresses all four response options, supporting reflection on why particular decisions are appropriate or inappropriate.
Scaffolded Support
The four lifelines offer optional assistance while allowing learners to decide when additional support is needed.
Game-Based Motivation
The progression ladder, virtual points, safety milestones, sound effects, and optional music establish a recognizable game structure.
Self-Paced Participation
The absence of a timer allows learners to consider questions and explanations without externally imposed time pressure.
These design decisions aim to balance engagement with opportunities for conceptual learning.
5.4 Initial Prototype: Conversational Quiz
Development began with an interactive conversational quiz derived from my coursework notes.
This initial activity allowed me to organize AI/ML questions and explanations into a practice-oriented format.
The experience helped establish the instructional content and question structure used in subsequent versions.
5.5 Second Prototype: Downloadable HTML Quiz
I expanded the conversational activity into a downloadable HTML resource containing 50 questions.
This version incorporated:
Topic-based question categories.
Hints and explanatory feedback.
Theme filters.
Short practice rounds.
Downloadable results.
This development stage transformed the original questions into a reusable, browser-based learning resource.
5.6 Third Prototype: Game-Show Edition
I proposed adapting the review resource into a game-show experience.
The subsequent development introduced a 15-level progression ladder, final-answer confirmation, four lifelines, virtual points, safety milestones, and synthesized audio.
The game was developed using HTML, CSS, and JavaScript, with browser-local storage for saving progress and browser-generated audio for sound effects.
ChatGPT supported coding, troubleshooting, and revision, while I directed the instructional requirements, question content, interaction design, and testing.
The game uses an original visual interface and synthesized audio. It is independently developed and is not affiliated with the television program that inspired its format.
5.7 Functional Testing and Iterative Refinement
I conducted developer play-testing to examine the game’s functionality, instructional interactions, feedback, audio features, and progress-saving behavior.
Testing prompted adjustments to the background-music volume and the addition of a volume slider, giving users greater control over the audio experience.
When the Brightspace learning management system appeared to restrict the interactive behavior of the uploaded HTML resource, I published a standalone web version to provide access through an external link.
This revision addressed a practical deployment limitation and improved the ability to share the resource outside the original course environment.
5.8 Testing Results and Observations
During a documented developer play-test, I completed all 15 questions correctly and reached the maximum score of 1,000,000 virtual points.
I used the Ask the Audience and Call a Coach lifelines on Question 14, which involved calculating classification accuracy.
A screenshot and downloadable learning record document this game session.
The play-test provided evidence that the game sequence could be completed and that the selected lifelines and learning-record features functioned during the test.
I also observed that the application recognized previously saved progress when reopened.
However, because I participated in the development, had prior exposure to the questions, and used lifeline assistance, the result cannot be interpreted as independent evidence of AI/ML mastery.
No formal evaluation of learning gains, interview readiness, or usability with the intended learner population has been completed.
The project combined AI-assisted development, instructional design, browser-based technologies, and web publication.
| Tool or Technology | Application in the Project |
|---|---|
| ChatGPT | Assisted with question development, explanatory feedback, coding, debugging, troubleshooting, and iterative refinement. |
| HTML | Structured the game interface, questions, feedback, and learner-facing components. |
| CSS | Supported the visual design, responsive layouts, progression ladder, and game-show presentation. |
| JavaScript | Implemented game logic, question selection, lifelines, scoring, feedback, and interaction controls. |
| Browser Local Storage | Enabled browser-based saving and resumption of game progress. |
| Web Audio Technologies | Supported synthesized sound effects, background music, and audio controls. |
| Coursework Notes and Quiz Materials | Provided the instructional foundation for question development and explanations. |
| Brightspace | Served as the initial course-based distribution environment before the standalone version was published. |
| Web Publication | Made the interactive game accessible through a standalone browser-based link. |
AI-Assisted Development Approach
I served as the individual learning designer and developer, directing ChatGPT to support the technical implementation of the game.
I provided the original instructional materials, identified the audience needs, specified the game mechanics, reviewed the questions and explanations, tested the application, and directed subsequent revisions.
AI-assisted tools supported implementation, but the instructional purpose, design requirements, and evaluation decisions remained under my direction.
The application does not require a live AI inference service to conduct the quiz and does not train a machine-learning model.
Who Wants to Be a Data Detective? is designed to provide learners with an engaging way to revisit foundational AI/ML concepts, apply knowledge to realistic situations, and identify areas requiring additional study.
For Beginners
The resource offers structured opportunities to practice essential concepts while receiving explanatory feedback and optional assistance.
For Graduate Students and Classmates
The game provides an alternative to conventional study notes and quizzes, accommodating learners with different academic and technical backgrounds.
For Aspiring AI Professionals
The scenario-based questions support review of terminology, data-processing decisions, model evaluation, and interpretation of results relevant to foundational technical discussions.
For Educators and Instructional Designers
The project demonstrates how existing course materials can be transformed into an interactive game-based learning experience using AI-assisted development.
Central Value Proposition
The project transforms foundational AI/ML coursework into an interactive, self-paced, and feedback-rich learning experience that combines practical scenarios, optional instructional support, and game-based engagement.
Its value lies in supporting opportunities for practice and reflection, not in certifying professional competence.
The project’s contribution lies in adapting familiar game-show mechanics to a purpose-built AI/ML learning experience.
Familiar Format Applied to Technical Learning
The game-show structure provides a recognizable framework for reviewing specialized concepts that may otherwise be presented through conventional quizzes.
Explanations for Every Response Option
Rather than presenting only the correct answer, the game provides explanations for all four alternatives.
This supports examination of common misconceptions and inappropriate decision-making strategies.
Support Through Optional Lifelines
Learners can request assistance without immediately receiving a complete explanation.
The simulated audience and coach functions introduce opportunities to consider additional information while retaining responsibility for the final answer.
Progression Across Learning Demands
The 15-question sequence progresses through foundations, application, and decision-making, encouraging learners to move beyond basic terminology.
Self-Paced Game-Based Learning
The game retains progression and reward mechanics without imposing a countdown timer.
Portable Learning and Progress Records
Browser-based progress saving and downloadable learning records support continuity and reflection.
Contextualized AI/ML Literacy
The question bank connects foundational concepts with education and industry scenarios, helping learners consider how technical decisions arise in real-world settings.
These features describe the project’s instructional design. They do not establish that the game is more effective than other educational review tools.
This project demonstrates my ability to identify a learning challenge, develop instructional content, apply game-based design principles, and create a functional educational application using AI-assisted development.
Audience Analysis
Identifying differences in learners’ prior knowledge and using those observations to inform the design of a supportive review resource.
Learning Objective Development
Translating foundational AI/ML coursework into specific learning goals involving terminology, problem recognition, method selection, and interpretation.
Instructional Content Development
Transforming coursework notes, quiz materials, and original scenario responses into a structured question bank.
Game-Based Learning Design
Incorporating progression, rewards, lifelines, feedback, and learner choice into an educational game.
Formative Assessment and Feedback Design
Developing explanations that help learners examine both correct and incorrect alternatives.
AI-Assisted Software Development
Using ChatGPT to support the implementation, troubleshooting, and refinement of a browser-based learning application.
Web Application Development
Applying HTML, CSS, JavaScript, browser storage, and audio functionality to deliver an interactive educational resource.
Testing and Iterative Refinement
Conducting developer play-tests, examining functionality, identifying interface limitations, and refining the application.
Evidence-Based Reporting
Documenting observed test results while distinguishing successful functional testing from unverified learning outcomes.
My Contribution
I conceived and developed Who Wants to Be a Data Detective? in response to learning needs I identified during my graduate residency.
I supplied the original instructional content, developed the learning objectives, organized the question categories, proposed the game-show adaptation, and directed the AI-assisted development process.
I also reviewed the questions and explanations, tested the game mechanics, requested interface and audio refinements, and published the standalone version.
The project demonstrates how my experience as an educator and learning designer informs the development of interactive educational technologies that address authentic learning needs.
Live Game Demonstration
Who Wants to Be a Data Detective?
https://adewole-data-detective.edtechcreator.chatgpt.site/
The application can be played directly in a web browser.
Supporting Project Evidence
The following artifacts document the project’s development and functionality:
Original Data Challenge coursework notes and scenario responses.
Initial conversational quiz.
Downloadable 50-question HTML quiz.
Game-show prototype with a 15-level progression ladder.
Screenshots showing the four lifelines.
Examples of explanatory feedback for response options.
Screenshots of audio controls and volume adjustments.
Screenshot of the completed 15-question developer play-test.
Downloadable learning record showing the completed run.
Evidence of browser-based progress saving and resumption.
Instructional and Technical Sources
My coursework notes, original Data Challenge responses, and course learning materials informed the development of the question bank and explanations.
The game-show format was adapted as an instructional design concept. The application uses independently developed presentation elements and is not affiliated with the original television program.
Current Status and Limitations
The browser-based educational game is functional and publicly accessible.
The documented developer play-test confirmed completion of the 15-question sequence and use of selected lifelines and reporting features.
However, the question difficulty bands are editorial estimates and have not undergone formal psychometric validation.
The simulated audience responses are not actual user votes and may be incorrect.
No independent study has established the game’s effects on learning gains, knowledge retention, interview readiness, or learner motivation.
Future development will prioritize gathering feedback from intended users on question clarity, difficulty, usability, and perceived usefulness. This feedback will inform further refinement of the questions and interface.