This project implements and evaluates a trigram language model using the NLTK Brown corpus. It performs next-word prediction with Laplace smoothing, sentence generation, probability estimation, and perplexity evaluation, demonstrating core concepts in statistical language modelling and NLP.
- Trigram language model implementation
- Laplace smoothing for stable predictions
- Next-word prediction (e.g. "I am β ...")
- Sentence generation
- Probability & perplexity evaluation
- Short story generation using Gemini API
- Python
- NLTK
- Jupyter Notebook
- Brown Corpus
N-Gram Language Model.ipynbβ implementation & experimentsN-Gram Language Model Report.pdfβ analysis & results