Update README.md

README.md

@@ -10,66 +10,55 @@ pinned: false
 license: cc-by-4.0
 ---
 
-# Stanza Parser with CoNLL-U Viewer
+# Greek Text Parser with spaCy displaCy
 
-A comprehensive linguistic analysis tool powered by Stanford's Stanza library that provides sentence parsing with multiple output formats.
+This Hugging Face Space provides a web interface for parsing Greek text using spaCy and visualizing the results in CoNLL-U format or as interactive dependency trees.
 
 ## Features
 
-- **Multi-language Support**: Parse text in English, Spanish, French, German, Chinese, Russian, and Arabic
-- **CoNLL-U Output**: Get standard linguistic annotation format output
-- **Interactive Data Table**: Browse parsed tokens with all linguistic features
-- **Dependency Visualization**: Text-based visualization of dependency relationships
-- **Copy-friendly Output**: Easy to copy results for use in other tools
+- **Greek Language Processing**: Uses spaCy's Greek language model (`el_core_news_sm`)
+- **CoNLL-U Output**: Generate standard CoNLL-U format for dependency parsing
+- **Visual Dependencies**: Interactive dependency tree visualization using displaCy
+- **Download Options**: Save HTML visualizations for offline viewing
 
-## What is CoNLL-U?
+## Usage
 
-CoNLL-U is a standard format for representing linguistic annotations that includes:
+1. Enter Greek text in the input field
+2. Choose your output format:
+   - **HTML**: Interactive dependency visualization
+   - **CoNLL-U**: Standard dependency parsing format
+3. Click "Parse Text" to process
 
-- **Tokenization**: Word and sentence boundaries
-- **Part-of-Speech Tagging**: Universal and language-specific POS tags
-- **Lemmatization**: Base forms of words
-- **Morphological Features**: Grammatical attributes
-- **Dependency Parsing**: Syntactic relationships between words
+## CoNLL-U Format
 
-## How to Use
+The CoNLL-U format includes:
+- Token ID
+- Word form
+- Lemma
+- Universal POS tag
+- Language-specific POS tag
+- Morphological features
+- Head token ID
+- Dependency relation
+- Enhanced dependencies
+- Miscellaneous annotations
 
-1. Enter your text in the input box
-2. Select the appropriate language
-3. Click "Parse Text" or press Enter
-4. View results in three formats:
-   - Raw CoNLL-U format (copy-paste ready)
-   - Interactive data table
-   - Dependency structure visualization
+## Examples
 
-## Example Output
-
-For the sentence "The cat sits on the mat", you'll get:
-
-- **CoNLL-U format**: Standard 10-column format with all linguistic features
-- **Data table**: Interactive view of each token's properties
-- **Dependencies**: "cat --nsubj--> sits", "mat --nmod--> sits", etc.
-
-## Use Cases
-
-- **Linguistic Research**: Analyze sentence structure and grammatical relationships
-- **NLP Development**: Generate training data or test parsing models
-- **Educational**: Learn about syntactic analysis and dependency grammar
-- **Text Processing**: Prepare annotated data for downstream tasks
+Try these Greek sentences:
+- `Ο γάτος κοιμάται στον καναπέ.` (The cat sleeps on the sofa)
+- `Η Μαρία διαβάζει ένα βιβλίο στη βιβλιοθήκη.` (Maria reads a book in the library)
+- `Τα παιδιά παίζουν ποδόσφαιρο στην αυλή.` (The children play football in the yard)
 
 ## Technical Details
 
-This space uses:
-- **Stanza**: Stanford's multilingual NLP toolkit
-- **Gradio**: For the interactive web interface
-- **Pandas**: For data table visualization
+- Built with Gradio for the web interface
+- Uses spaCy 3.4+ with Greek language support
+- Generates standard CoNLL-U format output
+- Provides downloadable HTML visualizations
 
-The models are automatically downloaded and cached when the space starts up.
-
-## Supported Languages
-
-Currently supports: English (en), Spanish (es), French (fr), German (de), Chinese (zh), Russian (ru), Arabic (ar)
-
----
+## Requirements
 
-*Powered by Stanford Stanza - https://stanfordnlp.github.io/stanza/*
+- Python 3.8+
+- spaCy with Greek model
+- Gradio for web interface