Personalized learning is increasingly gaining importance in the field of educational technology, particularly

among young learners who require interactive tools and personalized content based on their learning speeds.

Most learning platforms currently available offer identical content to all learners irrespective of differences in

their comprehension, attention, and revision capabilities. This paper introduces an innovative AI Tutor designed

for students in classes 1 to 5 by integrating adaptive learning, quizzes, speech communication, and analytical

efficient topic tracking, and valuable personalization suggestions for early learners.

Keywords— Personalized learning, artificial intelligence, adaptive learning, educational technology, Firebase,

speech recognition, recommendation system, AI tutor.

The rapid growth of digital learning platforms has transformed the way students access educational content and

interactive learning resources. However, many traditional e-learning systems still provide the same educational

material to all students without considering individual learning capability, learning speed, or understanding level.

This often reduces student engagement and limits personalized learning support for beginner-level learners.

To address this problem, the proposed “Personalized AI Tutor” introduces an adaptive learning platform

designed to provide personalized educational interaction for early learners. The system combines image-based

The proposed system is developed using HTML, CSS, JavaScript, Firebase Authentication, Firebase Cloud

Firestore, Chart.js, and browser-based speech technologies. Educational modules including alphabets, numbers,

shapes, colors, animals, fruits, vehicles, objects, and phonics learning are integrated to support interactive

educational learning for beginner-level students.

Overall, the Personalized AI Tutor provides an intelligent and scalable learning environment capable of

improving adaptive learning support and interactive learning experiences using modern web technologies.

Recent advancements in Artificial Intelligence and adaptive learning technologies have significantly improved

intelligent tutoring systems and personalized educational platforms. Modern AI-based educational systems use

recommendation mechanisms, learner analytics, adaptive quizzes, and conversational interaction to improve

However, speech-assisted interaction for beginner-level learners was not included.

Lin et al. [2] reviewed Artificial Intelligence-based intelligent tutoring systems designed for adaptive and

sustainable education. Their study demonstrated how adaptive learning technologies improve learner

engagement and personalized educational support. However, most systems focused mainly on theoretical

analysis rather than real-time learner analytics and dashboard implementation.

Adiguzel et al. [3] explored the impact of ChatGPT and Generative Artificial Intelligence technologies in modern

education systems. Their work demonstrated that conversational AI improves learner interaction, accessibility,

and personalized educational support. However, the study mainly focused on conversational tutoring and did not

include adaptive topic unlocking, learner analytics, or quiz-based recommendation mechanisms.

Contrino et al. [4] investigated adaptive learning tools and their impact on student performance and educational

higher-level learners and did not include phonics-based learning or adaptive revision support for beginner-level

students.

Although existing intelligent tutoring systems support adaptive learning and conversational educational

interaction, many platforms still lack integrated personalized learning support for beginner-level students. Most

systems focus only on chatbot interaction, recommendation systems, or learner analytics independently rather

than combining all features into a unified adaptive learning environment.

Current educational platforms often lack real-time weak-area analysis, adaptive topic unlocking, learner

analytics dashboards, speech-assisted interaction, and AI-based quiz recommendation mechanisms. Many

systems are also designed mainly for higher-level students instead of early learners.

Additionally, several existing platforms lack scalable cloud-based storage and real-time learner performance

monitoring. Learning analytics such as topic mastery, learning consistency, weak-area tracking, and adaptive

revision support are not effectively integrated into many current systems.

recommended.

Speech-assisted interaction and Text-to-Speech technologies are integrated to improve learner engagement and

pronunciation support for beginner-level students. Adaptive quizzes are generated according to topic progress

and student understanding levels.

Firebase Cloud Firestore is used for secure real-time storage of learner analytics, quiz results, recommendation

history, and educational progress tracking. The platform also includes dashboard analytics and radar-chart-based

learner performance visualization for monitoring educational growth and topic mastery.

The proposed Personalized AI Tutor system follows an adaptive learning methodology designed for beginner-

level students. The platform combines personalized educational content, AI-based quiz evaluation, learner

Once learning activities are completed, the system generates adaptive quizzes related to the selected topic. Quiz

scores, topic completion status, learning consistency, and weak-area analysis are continuously monitored and

stored inside Firebase Cloud Firestore for learner analytics and recommendation generation.

The adaptive recommendation engine analyzes learner performance dynamically. If low scores or repeated

mistakes are detected, the system recommends revision activities and additional practice questions for the same

topic.

The platform also integrates learner analytics dashboards and radar chart visualizations to monitor student

performance, educational progress, topic mastery, and learning consistency. This adaptive learning approach

improves personalized learning support and learner engagement for early-level students.

Fig.1. System Architecture

The Learning Module provides image-assisted educational content for topics including alphabets, numbers,

phonics, colors, shapes, animals, fruits, vehicles, and objects. Text-to-Speech interaction is integrated to improve

learner engagement and pronunciation support.

The Quiz Module dynamically generates topic-based educational assessments according to student progress.

Quiz performance is analyzed to identify weak learning areas and recommendation requirements.

The adaptive learning workflow therefore helps students improve weak learning areas gradually while

maintaining personalized learning progression according to individual learning capability.

Several performance evaluation metrics are used to monitor learner progress and system effectiveness within the

Quiz Accuracy measures the percentage of correct answers provided by students during topic-based assessments.

Higher quiz accuracy indicates improved topic understanding and learning performance.

receive personalized educational guidance and progressive topic advancement according to their current learning

capability and educational performance.

The Personalized AI Tutor system was implemented using HTML, CSS, JavaScript, Firebase Authentication,

Firebase Cloud Firestore, Chart.js, and browser-based speech technologies.

Educational datasets containing beginner-level learning topics such as alphabets, numbers, phonics, shapes,

colors, animals, fruits, vehicles, and objects were integrated into the platform. Adaptive quizzes, learner

analytics, recommendation mechanisms, and speech-assisted interaction modules were also implemented.

Student quiz records, topic mastery, learning consistency, weak-area analysis, and educational progress were

continuously stored and monitored using Firebase Cloud Firestore. Radar charts and educational dashboards

System testing was performed across educational modules including adaptive quizzes, recommendation

generation, phonics learning, dashboard analytics, and learner progress tracking. The results demonstrated stable

adaptive learning performance and improved learner engagement.

The adaptive recommendation engine dynamically identifies weak educational areas and recommends suitable

revision activities and personalized practice modules according to student quiz performance and learning

progress.

Fig.2. Personalized AI Tutor Dashboard

The Personalized AI Tutor dashboard provides real-time learner analytics using Firebase Cloud Firestore

synchronization. Student progress, topic mastery, quiz accuracy, and learning consistency are continuously

monitored and visualized using interactive dashboard analytics and radar charts.

Fig.3. Topic Strength Radar Analysis

The radar chart analyzes quiz accuracy, learning progress, topic mastery, educational consistency, adaptive

learning score, and learner engagement across multiple educational modules. The analytics help identify both

strong and weak learning areas dynamically.

The adaptive learning engine continuously updates learner profiles using Firebase Cloud Firestore and generates

personalized educational recommendations according to student performance and learning consistency.

The AI Learning Profile therefore helps monitor learner growth, adaptive learning progression, and topic-wise

performance improvement in real time.