Repository navigation
Installation Instructions for the NLP Suite
The NLP Suite is distributed as a standalone application — no Python, Anaconda, or Java installation required.
- Go to Releases
- Download the latest zip for your platform:
-
NLP-Suite-windows-x64(Windows 10/11) -
NLP-Suite-mac-arm64(macOS Apple Silicon — M1/M2/M3/M4/M5)
-
No GitHub account needed.
| Resource | Minimum |
|---|---|
| Free disk space | 5 GB minimum; 8 GB with BERT features; ~10 GB with all optional software (see breakdown below) |
| Internet | Required for first run (model downloads) and for optional features (geocoding, web scraping) |
| RAM | 4 GB (8 GB recommended for large corpora) |
| Component | Size | When downloaded |
|---|---|---|
| Installer zip | ~1.5–2 GB | Initial download |
| Extracted application | ~3.5 GB | On extraction |
| Stanza English model | ~525 MB | First time you run a Stanza analysis |
| spaCy English model | ~50 MB | First time you run a spaCy analysis |
| NLTK data (wordnet, punkt, etc.) | Bundled | Already included in the installer |
| BERT models (NER, similarity, topic modeling, WSI) | Up to ~3 GB | Only if you use BERT features, downloaded on first use of each |
| MALLET (topic modeling) | ~50 MB | Only if you use MALLET; external download (see below) |
| Stanford CoreNLP | ~550 MB | Only if you use CoreNLP instead of Stanza; requires Java |
| Java JDK | ~300 MB | Only if you use Stanford CoreNLP |
| Google Earth Pro | ~50 MB | Only if you use KML/Google Earth visualizations; Folium (included) is an alternative |
- Windows: Right-click the downloaded zip → Extract All
- macOS: Double-click the zip in Finder
Double-click NLP_Suite.exe in the extracted folder.
If Windows shows "Windows protected your PC":
- Click More info
- Click Run anyway
This only happens the first time.
The download includes NLP Suite Mac Setup.app — use it to launch the NLP Suite for the first time:
- Double-click NLP Suite Mac Setup
- Click Select NLP Suite Folder and point it to the extracted
NLP_Suitefolder - Click Remove Quarantine from Executable — this clears the macOS Gatekeeper block
- Click Remove Quarantine from Folder — allows all bundled files to run
- Click Open NLP Suite
After the first launch, you can also run NLP_Suite directly by double-clicking it.
Why is this needed? macOS blocks unsigned applications downloaded from the internet (Gatekeeper). The Setup app removes this quarantine flag so the NLP Suite can run. This is a one-time step.
On the very first launch, the NLP Suite automatically creates two default configuration files so you can start testing right away:
| Setting | Default value |
|---|---|
| Input folder |
lib/sampleData/newspaperArticles (bundled sample texts) |
| Output folder |
~/Documents/NLP_output (your Documents folder) |
| NLP package | Stanza |
| Language | English |
You can change these at any time via the I/O Configuration options at the top of the main window.
The standalone distribution bundles everything the NLP Suite needs:
- Python runtime and all Python packages (Stanza, spaCy, NLTK, BERT, Plotly, matplotlib, etc.)
- WordNet lexical database (via NLTK — no separate download needed)
- NRC Emotion Lexicon for emotion analysis
-
Sample data for testing (
lib/sampleData/) - TIPS documentation and reminders
| Tool | When needed | What to do |
|---|---|---|
| Stanford CoreNLP | Only if you select CoreNLP as your NLP package | Download from Stanford — but Stanza (included) does the same things without Java |
| Java | Only if you use Stanford CoreNLP | Download JDK |
| Google Earth Pro | Only for KML/Google Earth map visualizations | Download — or use Folium (included) as an alternative |
| Google Maps API key | Only for Google Maps visualizations | Get an API key — or use Folium (included) |
| Gephi | Only for network graph visualization | Download |
Tip: For most users, the bundled tools (Stanza, Folium, vis.js) cover all functionality without any additional downloads.
If you want to modify the NLP Suite source code or contribute to development:
# Clone the repository
git clone https://github.com/NLP-Suite/NLP-Suite.git
cd NLP-Suite
# Create a conda environment
conda create -n NLP python=3.10 -y
conda activate NLP
# Install dependencies
pip install pandas numpy scipy scikit-learn openpyxl xlrd
pip install stanza spacy nltk gensim textblob textstat nrclex
pip install langdetect langid autocorrect pyspellchecker fuzzywuzzy
pip install transformers sentence-transformers
pip install matplotlib plotly seaborn wordcloud mpld3 kaleido
pip install folium geopy simplekml shapely geopandas
pip install requests beautifulsoup4 lxml chardet tqdm psutil
pip install pdfminer.six python-docx striprtf
pip install Pillow SPARQLWrapper
# Run the NLP Suite
python NLP_Suite_main.py| Branch | Purpose |
|---|---|
roberto |
Active development — latest features |
current-stable |
Stable release — mirrors roberto after testing |
| Problem | Solution |
|---|---|
| Windows: "Windows protected your PC" (SmartScreen) | Right-click NLP_Suite.exe → Properties → click Unlock (bottom right) → OK or Apply. (If that doesn't appear, try clicking More info → Run anyway) |
| macOS: "cannot be opened because it is from an unidentified developer" | Use the NLP Suite Mac Setup app to remove quarantine, then click Open NLP Suite |
| macOS: "NLP_Suite is damaged and can't be opened" | Use the Mac Setup app to remove quarantine. If that doesn't work, open Terminal and run: xattr -cr /path/to/NLP_Suite
|
| First run: no config files found | Normal — the app creates them automatically. If it doesn't, check that your Documents folder is writable |
| Stanza model download fails | Check your internet connection. Stanza downloads language models on first use (~525 MB for English) |
- Click the ? buttons next to each option in any GUI for explanations
- Use the TIPS and README buttons at the bottom of each GUI
- Visit the Troubleshooting wiki page
- Report an issue
Visit the NLP Suite Architecture page for information on the NLP Suite design.
){kind=logo}