markgewhite commited on
Commit
aaac9ac
Β·
1 Parent(s): 66c4727

Updated README

Browse files
Files changed (1) hide show
  1. README.md +9 -5
README.md CHANGED
@@ -19,8 +19,10 @@ Track how research concepts evolve over time using AI-powered semantic embedding
19
  - πŸ“Š **Track concept evolution** through time windows
20
  - 🧠 **Semantic embeddings** using sentence-transformers (MiniLM-L6-v2)
21
  - πŸ“ˆ **Interactive timeline** visualization with Gradio
 
22
  - 🎯 **Kalman filter** constraints prevent unrealistic concept jumps
23
  - πŸ’Ύ **Local embeddings cache** - no API costs
 
24
  - 🌐 **Works locally and on HuggingFace Spaces**
25
 
26
  ## Quick Start
@@ -103,9 +105,10 @@ Your app will be live at: `https://huggingface.co/spaces/YOUR-USERNAME/concept-t
103
 
104
  ### 3. View Results
105
 
106
- - Timeline showing concept evolution across time windows
107
- - Papers ranked by similarity to concept
108
- - Statistics: high/moderate/low confidence counts
 
109
 
110
  ## How It Works
111
 
@@ -182,11 +185,12 @@ pytest tests/test_arxiv_client.py -v
182
  concept_tracker/
183
  β”œβ”€β”€ app.py # Gradio interface
184
  β”œβ”€β”€ backend/
185
- β”‚ β”œβ”€β”€ gradio_wrapper.py # Gradio event handlers
186
  β”‚ β”œβ”€β”€ tracker.py # Main orchestrator
187
  β”‚ β”œβ”€β”€ arxiv_client.py # ArXiv API wrapper
188
  β”‚ β”œβ”€β”€ embedding_service.py # Embeddings + cache
189
  β”‚ β”œβ”€β”€ kalman_tracker.py # Kalman filter logic
 
190
  β”‚ β”œβ”€β”€ config.py # Configuration
191
  β”‚ β”œβ”€β”€ models.py # Pydantic data models
192
  β”‚ └── utils/
@@ -254,7 +258,7 @@ concept_tracker/
254
  - πŸ”² **Tree branching** with HDBSCAN clustering
255
  - πŸ”² **Bidirectional tracking** (trace concepts to origins)
256
  - πŸ”² **Multi-signal validation** (citations, author overlap)
257
- - πŸ”² **Export results** to JSON/CSV
258
 
259
  ## License
260
 
 
19
  - πŸ“Š **Track concept evolution** through time windows
20
  - 🧠 **Semantic embeddings** using sentence-transformers (MiniLM-L6-v2)
21
  - πŸ“ˆ **Interactive timeline** visualization with Gradio
22
+ - πŸ—ΊοΈ **t-SNE visualization** showing concept trajectory through embedding space
23
  - 🎯 **Kalman filter** constraints prevent unrealistic concept jumps
24
  - πŸ’Ύ **Local embeddings cache** - no API costs
25
+ - πŸ“₯ **Export results** to JSON or CSV
26
  - 🌐 **Works locally and on HuggingFace Spaces**
27
 
28
  ## Quick Start
 
105
 
106
  ### 3. View Results
107
 
108
+ - **Timeline**: Concept evolution across time windows with expandable paper lists
109
+ - **Visualization**: t-SNE plot showing concept trajectory through embedding space
110
+ - **Statistics**: Total papers, time steps, and high/moderate/low confidence counts
111
+ - **Export**: Download results as JSON (full data) or CSV (flattened paper records)
112
 
113
  ## How It Works
114
 
 
185
  concept_tracker/
186
  β”œβ”€β”€ app.py # Gradio interface
187
  β”œβ”€β”€ backend/
188
+ β”‚ β”œβ”€β”€ gradio_wrapper.py # Gradio event handlers + export
189
  β”‚ β”œβ”€β”€ tracker.py # Main orchestrator
190
  β”‚ β”œβ”€β”€ arxiv_client.py # ArXiv API wrapper
191
  β”‚ β”œβ”€β”€ embedding_service.py # Embeddings + cache
192
  β”‚ β”œβ”€β”€ kalman_tracker.py # Kalman filter logic
193
+ β”‚ β”œβ”€β”€ visualization.py # t-SNE visualization
194
  β”‚ β”œβ”€β”€ config.py # Configuration
195
  β”‚ β”œβ”€β”€ models.py # Pydantic data models
196
  β”‚ └── utils/
 
258
  - πŸ”² **Tree branching** with HDBSCAN clustering
259
  - πŸ”² **Bidirectional tracking** (trace concepts to origins)
260
  - πŸ”² **Multi-signal validation** (citations, author overlap)
261
+ - βœ… **Export results** to JSON/CSV
262
 
263
  ## License
264