Chatbot Analytics Accelerator
Session-level chatbot conversation analytics from a silver layer: resolution rate, satisfaction, escalation, and keyword topic classification into gold KPI tables. Ships a dependency-free analytics library + offline tests. Fully generic Databricks Asset Bundle.
⚙ Try the Live Demo interactive
Enter session outcomes and see resolution rate, escalation, CSAT and topic mix — the accelerator's KPI logic, live.
⚡ Open KPI Calculator📋 What's Inside 16 files
- ✓ README.md
- ✓ LICENSE
- ✓ manifest.json
- ✓ databricks.yml
- ✓ resources/jobs.yml
- ✓ src/silver_to_gold.py
- ✓ src/topic_classification.py
- ✓ src/session_metrics.sql
- ✓ lib/session_metrics.py
- ✓ tests/test_session_metrics.py
- ✓ conftest.py
- ✓ guide/01_what-you-get.md
- ✓ guide/02_getting-started.md
- ✓ guide/03_architecture.md
- ✓ guide/04_support.md
- ✓ guides/chatbot-analytics-methodology.md
📁 File Structure 16 files
├── README.md
├── LICENSE
├── manifest.json
├── databricks.yml
├── resources/
│ ├── jobs.yml
├── src/
│ ├── silver_to_gold.py
│ ├── topic_classification.py
│ ├── session_metrics.sql
├── lib/
│ ├── session_metrics.py
├── tests/
│ ├── test_session_metrics.py
├── conftest.py
├── guide/
│ ├── 01_what-you-get.md
│ ├── 02_getting-started.md
│ ├── 03_architecture.md
│ ├── 04_support.md
├── guides/
│ ├── chatbot-analytics-methodology.md
📖 Documentation Preview README excerpt
Chatbot Analytics Accelerator
Session-level **chatbot conversation analytics** from a silver layer: resolution
rate, satisfaction (CSAT), escalation rate, and topic classification, published
to gold tables for a monitoring dashboard.
Fully generic and environment-agnostic: point it at your silver conversation
data and deploy.
What's inside
- **Databricks Asset Bundle** — a daily silver→gold transform + KPI aggregation.
- **Session KPIs** — resolution, escalation, CSAT, and message counts per topic
per day.
- **Keyword topic classification** — billing / outage / contract / metering /
technical / other.
- **A dependency-free analytics library** (`lib/session_metrics.py`): resolution/
escalation rates, CSAT, message averages, topic classification/distribution,
and a KPI rollup — with an offline test suite (6 tests).
Quickstart
```bash
pip install pytest && pytest tests/ -v
databricks bundle deploy -t dev -p <profile>
```
Provide silver conversation data; the transform derives the gold sessions table.
License
MIT — see `LICENSE`.
... preview truncated, see full README in product download.