Uhlelo Lokuhlaziya Ikhophasi YesiZulu | IsiZulu Corpus Analysis Toolkit
# π¬ Uhlelo Lokuhlaziya Ikhophasi YesiZulu
## IsiZulu Corpus Analysis Toolkit
A comprehensive web-based corpus analysis toolkit specifically designed for IsiZulu linguistic research. This application provides advanced computational linguistics tools for analyzing IsiZulu text corpora, featuring both traditional corpus analysis methods and modern visualization techniques.
## π Academic Context
This project was developed as part of PhD Linguistics research focusing on IsiZulu language analysis. It serves as a practical tool for computational linguistics research, corpus studies, and linguistic data analysis within the context of African language processing.
**Authors:**
- **Mr. Mthuli Percival Buthelezi** π - PhD Linguistics [IsiZulu] Candidate
- **Mr. Sakhile Marcus Zungu** π - MSc Applied Mathematics [Astronomy] Candidate
- **Research Collaboration:** Uluntu Algorithms
## β¨ Features
### Core Analysis Tools
| Feature | IsiZulu Name | Description |
|---------|--------------|-------------|
| π **Word Frequency** | *Uhlumagama* | Statistical analysis of word occurrence with frequency distribution |
| π **Concordance** | *Imvumelwanomagama* | Contextual word search showing surrounding text |
| π― **Keyness Analysis** | *Ubungqikithimagama* | Comparative analysis against reference corpus using log-likelihood |
| π **N-grams Analysis** | *Onhlamvunye* | Extract uni-grams, bi-grams, tri-grams, and higher-order sequences |
| βοΈ **Word Cloud** | *Amafumagama* | Visual representation of word frequency distribution |
| π€ **Letter Frequency** | *Isibalo sezinhlamvu zamagama* | Character-level statistical analysis |
| π **Word Count** | *Isibalo samagama* | Basic corpus statistics and metrics |
### Technical Capabilities
- **Multi-encoding Support**: Automatic encoding detection (UTF-8, Latin-1, ANSI)
- **Interactive Web Interface**: Built with Streamlit for easy deployment
- **Data Export**: Download results as CSV files
- **Visualization**: Matplotlib-based word clo β¦