NLP
Prizm AI
Progressive Hint & Error Detection System
Created hybrid rule-based and ML system for e-learning platform boosting course completion by 20%. Engineered with scikit-learn and spaCy achieving 95% accuracy across 5k+ coding exercises. Integrated BERT-based sentiment analysis for feedback prioritization, improving learner satisfaction by 25%.
5 months
Project Duration
95% accuracy
Key Achievement
4+
Technologies Used
The Challenge
Prizm AI's e-learning platform needed an intelligent system to provide progressive hints and detect errors in real-time to improve student engagement and learning outcomes.
The Solution
Created a hybrid ML system combining scikit-learn algorithms with spaCy NLP processing and BERT fine-tuning for contextual understanding of student responses and intelligent hint generation.
Results & Impact
- Achieved 95% accuracy in error detection
- Increased course completion rates by 20%
- Improved student satisfaction scores by 25%
- Reduced instructor workload by 40%
Project Metrics
95% accuracy
20% course completion boost
25% satisfaction improvement
Technologies Used
scikit-learnspaCyBERTPython
Project Details
Company
Prizm AI
Duration
5 months
Team Size
3 engineers + 2 education specialists
Category
NLP