TLDR: AnveshanaAI is an innovative AI/ML learning platform that offers a personalized, gamified, and adaptive educational experience. It features automated question generation, interactive assessment modes (including simulations and AI-powered oral exams), and a robust dataset aligned with Bloom’s taxonomy. The platform aims to address the limitations of static learning environments by providing dynamic challenges, continuous progress tracking, and explainable AI techniques to enhance learner engagement and understanding.
The rapid evolution of artificial intelligence (AI) and machine learning (ML) has created a significant demand for effective learning platforms. While many existing coding environments offer structured exercises, they often fall short in addressing the unique requirements of AI/ML education, which demands not just algorithmic implementation but also conceptual reasoning, model experimentation, and result interpretation.
Addressing these gaps, researchers have proposed AnveshanaAI, an innovative application-based learning platform designed to transform AI/ML education. This platform moves beyond traditional static question repositories by integrating adaptive, automated, and domain-aware assessment methods, all grounded in Bloom’s taxonomy for a balanced learning progression. The goal is to provide a comprehensive and engaging ecosystem that unifies automated question generation, adaptive assessment, and validation mechanisms.
A Personalized and Gamified Learning Experience
AnveshanaAI offers learners a highly personalized experience, starting with a dynamic dashboard that tracks progress through streaks, levels, and badges. This gamified approach is designed to enhance engagement and motivation. The platform provides structured navigation across diverse AI/ML domains, including data science, machine learning, deep learning, transformers, generative AI, large language models, and multimodal AI, with the flexibility to expand further.
Learners can switch between various interactive modes such as the Playground for coding, Challenges for problem-solving, a Simulator for hands-on experimentation with models, a Dashboard for progress tracking, and a Community for collaboration. This multi-modal approach supports exploration and practical application of concepts.
Automated Question Generation and Adaptive Assessment
A core innovation of AnveshanaAI is its ability to automatically generate diverse and pedagogically meaningful challenges. Unlike platforms relying on fixed question sets, AnveshanaAI ensures a balanced learning progression. It incorporates semantic similarity checks and explainable AI techniques to improve transparency and reliability in its assessment methods.
The platform’s underlying dataset is meticulously constructed with over 10,000 problem-answer pairs across various AI/ML domains. Each task is enriched with metadata, including Bloom’s taxonomy level and difficulty annotations, enabling adaptive delivery and personalized learning paths. The dataset also employs augmentation strategies like difficulty scaling and cross-mode adaptation, transforming base problems into coding, simulation, debugging, and even viva-style formats to expand task variety.
Robust Architecture and Methodology
AnveshanaAI’s system architecture is modular, featuring a Question Generation Pipeline (QGP), an adaptive delivery engine, a multimodal interaction layer, and an analytics dashboard. It leverages fine-tuned Large Language Models (LLMs) for question generation, Docker-based sandboxes for secure code execution, and vector databases for efficient context retrieval. A notable feature is the Viva Mode, which uses Whisper-based Automatic Speech Recognition (ASR) for oral-style question-and-answer sessions, assessing conceptual clarity and reasoning.
The platform’s frontend is built with React and Vite for responsiveness, while the backend uses Node.js and Express, supported by a MySQL database for secure data storage. User performance is continuously tracked across all activities, and this data is processed using machine learning techniques to generate personalized feedback, adjust difficulty, and recommend further learning, ensuring an adaptive growth experience.
Also Read:
- AI-Powered COTUTOR Model Optimizes Personalized Learning for Programmers
- CLPO: A Self-Evolving Learning Approach for Enhanced LLM Reasoning
Validation and Future Impact
Experiments conducted on AnveshanaAI demonstrate broad dataset coverage, stable fine-tuning of models (like Mistral 7B) with reduced perplexity, and measurable gains in learner engagement. The explainable AI analysis confirms that the model’s predictions are guided by contextually relevant information, enhancing interpretability and transparency.
In conclusion, AnveshanaAI integrates adaptivity, gamification, interactivity, and explainability to support next-generation AI education. It provides a reliable dataset and an effective platform for transparent, adaptive, and educationally aligned AI systems. For more details, you can refer to the full research paper.
