After completing this experiment, students will be able to:

1. Build a Chunker: Implement chunking using HMM and CRF models for NLP tasks.
2. Experiment with Features: Analyze how different feature sets (lexicon, POS tags) affect chunking accuracy.
3. Evaluate Corpus Size Impact: Assess how training data size influences model performance.
4. Visualize and Interpret Results: Use simulation outputs to compare chunking accuracy and error patterns.
5. Apply Chunking Knowledge: Understand the role of chunking in downstream NLP applications (information extraction, parsing, etc.).

Learning Focus

• Construct chunkers using different algorithms and features
• Compare chunking accuracy across configurations
• Apply chunking principles to real linguistic data