An AI-powered log analysis dashboard that helps engineers detect anomalies, group similar errors, and identify root causes from system logs.
https://log-intelligence-app-l6mqdpuchvmpcptzjnfxqr.streamlit.app/
-
๐ Semantic Log Understanding Uses transformer embeddings to understand log meaning beyond keywords
-
๐ Log Clustering Groups similar logs using KMeans for faster debugging
-
๐จ Anomaly Detection Detects unusual logs using Isolation Forest
-
๐ Time-based Trend Analysis Identifies spikes and patterns over time
-
๐ค Root Cause Analysis Highlights dominant error clusters causing system issues
- Python
- Streamlit
- FAISS (vector similarity search)
- Sentence Transformers
- Scikit-learn (KMeans, Isolation Forest)
- Pandas & NumPy
-
Upload a log file (.txt)
-
Logs are converted into embeddings using a transformer model
-
System performs:
- Clustering (KMeans)
- Anomaly detection (Isolation Forest)
-
Results are visualized in an interactive dashboard
-
Root cause is inferred from dominant anomaly clusters
log-intelligence-app/
โโโ app.py
โโโ requirements.txt
โโโ README.md
pip install -r requirements.txt
streamlit run app.pyThis app is deployed using Streamlit Community Cloud.
Steps:
- Push code to GitHub
- Connect repository to Streamlit Cloud
- Deploy using
app.pyas entry point
2024-01-01 10:00:00 INFO User login successful
2024-01-01 10:05:00 ERROR Database connection timeout
2024-01-01 10:10:00 WARNING High memory usage detected
Built a production-style log intelligence system that clusters logs using KMeans on embeddings, detects anomalies via Isolation Forest, and performs root cause analysis with time-based trend visualization.
- Real-time log streaming
- LLM-based debugging assistant
- Advanced clustering (DBSCAN / HDBSCAN)
- Docker deployment
This project is for educational and demonstration purposes.