How bhoomikakm04/interactive-lda-topic-explorer works
This is a Streamlit dashboard for exploring LDA topics in a document collection. Users supply a CSV or use the bundled corpus, inspect preprocessing and document-term representations, configure and run topic modeling, then review topic-word lists, interpretation reports, and visualizations. The code is concentrated in app.py; the sampled excerpts show input validation, preprocessing helpers, and vectorization, while several later dashboard sections are documented but not excerpted. Dashed edges mark workflow links whose implementation details were not shown.
Main entry points: app.py Streamlit Dashboard · lda_corpus.csv Default Corpus
Components 11
Corpus Input
- Default Corpus
data/lda_corpus.csv - CSV Upload
app.py - Input Validation
app.py
Topic Modeling
- Text Preprocessing
app.py - Document-Term Matrix
app.py - LDA Model
app.py
Topic Results
- Topic Word Lists
app.py - Interpretation Reports
app.py - Interactive Visualizations
app.py
Dashboard
- Streamlit Dashboard
app.py
Other components
- Dashboard User
Connections 15
- Dashboard User → Streamlit Dashboard: uses
README.md - Default Corpus → Input Validation: loads by default
- CSV Upload → Input Validation: validates upload
app.py - Streamlit Dashboard → CSV Upload: accepts upload
app.py - Input Validation → Text Preprocessing: prepares corpus
app.py - Text Preprocessing → Document-Term Matrix: vectorizes text
app.py - Document-Term Matrix → LDA Model: models counts
app.py - LDA Model → Topic Word Lists: produces topics
README.md - LDA Model → Interpretation Reports: informs report
README.md - LDA Model → Interactive Visualizations: feeds charts
README.md - Streamlit Dashboard → Text Preprocessing: shows preprocessing
README.md - Streamlit Dashboard → Document-Term Matrix: shows matrix
README.md - Streamlit Dashboard → Topic Word Lists: shows topics
README.md - Streamlit Dashboard → Interpretation Reports: shows report
README.md - Streamlit Dashboard → Interactive Visualizations: shows charts
README.md