snickerdust/PLN-summary

chore: include firebase credentials for Hugging Face Space

fix: Restore Hugging Face Spaces YAML metadata in README

docs: Update README.md with project overview and features

chore: remove thesis drafts and documentation from repository, keeping them local

chore: untrack firebase credentials and update gitignore

chore: setup gitignore and add documentation files

Update logo to CurricuLens (PLN Theme) and clean up SUS from questionnaire banner

Tambah box asal program studi, fakultas, universitas pada form testing

feat: add delete document button to sidebar and update black box testing evaluation form with search bar and delete scenarios

feat: increase max file upload limit to 100MB

feat: add LibreOffice support in Dockerfile and simplify extractor.py for legacy doc/ppt conversion

Ignore indobert_finetuned and venv folders in firebase.json to prevent large file uploads to Firebase Hosting

Convert tumpang tindih sentence similarity to pure Jaccard Index, removing all TF-IDF and Cosine dependencies from overlap calculation

Update main overlap_score calculation to use pure Jaccard Index of stopword-filtered keyword sets

Implement Jaccard index stopword and conjunction filtering in compute_overlap function

Restore weighted ranking selection in BERT backend while preserving raw cosine similarity scores for frontend UI display

Clean up project structure: delete model_lexical.py and remove all its references/fallbacks, move compute_overlap to preprocessor.py, and remove TEKS COCOK badge from sidebar, retaining only sentence snippet previews

Update index.html to render search snippet previews containing matching original sentences under document items in sidebar

Fix summary score calculation (remove source_weight multiplication) and add real-time search filtering & text-matching in the sidebar & summary panels

feat: sort summary by score descending in frontend display

fix: upgrade torch to 2.5.1+cpu — fixes torch.distributed.tensor.device_mesh import error with latest transformers

fix: simplify topic name to 'Topik N', add bert_error field to API result for debugging

fix: upgrade torch to 2.4.0+cpu (required by sentence-transformers>=3) and pin numpy<2 for binary compat

perf: use torch CPU-only wheel (~200MB vs 1.5GB) to speed up HF Spaces rebuild

refactor: simplify BERTopic pipeline to exact LOTO config from notebook (no fallback scenarios)

chore: align model print statement with BERT model backbone

fix: implement robust BERTopic fallback search to resolve empty topic results and 2.0 scores

Fix scipy linalg eigh crash on small datasets by implementing dynamic UMAP/HDBSCAN parameter scaling

Remove model weights and duplicate clone from Git tracking, add to gitignore

Update model to snickerdust/FT-PLN-IndoBERT and apply optimal LOTO parameters

add BERTopic support for topic modeling

remove ipynb files

Merge branch 'main' of https://huggingface.co/spaces/snickerdust/PLN-summary

upload ke HF

initial commit