{"id":28059552,"url":"https://github.com/moses000/mysoftware-nocnetintel","last_synced_at":"2026-05-05T21:31:52.660Z","repository":{"id":292115839,"uuid":"979865499","full_name":"moses000/mysoftware-nocNetIntel","owner":"moses000","description":"AI-powered NOC assistant for forecasting network outages, analyzing root causes, and recommending proactive resolutions using LLM and time-series intelligence.","archived":false,"fork":false,"pushed_at":"2025-05-08T07:55:57.000Z","size":22,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-05-12T08:06:55.145Z","etag":null,"topics":["ai","data-pipeline","deep-learning","docker","forecasting","llm","mlops","network-operations","nlp","noc","postgresql","predictive-maintenance","pytorch","root-cause-analysis","telecom","time-series","unsupervised-learning"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/moses000.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-05-08T07:32:06.000Z","updated_at":"2025-05-08T07:56:00.000Z","dependencies_parsed_at":"2025-05-08T08:47:26.446Z","dependency_job_id":null,"html_url":"https://github.com/moses000/mysoftware-nocNetIntel","commit_stats":null,"previous_names":["moses000/mysoftware-nocnetintel"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/moses000%2Fmysoftware-nocNetIntel","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/moses000%2Fmysoftware-nocNetIntel/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/moses000%2Fmysoftware-nocNetIntel/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/moses000%2Fmysoftware-nocNetIntel/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/moses000","download_url":"https://codeload.github.com/moses000/mysoftware-nocNetIntel/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253700618,"owners_count":21949694,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","data-pipeline","deep-learning","docker","forecasting","llm","mlops","network-operations","nlp","noc","postgresql","predictive-maintenance","pytorch","root-cause-analysis","telecom","time-series","unsupervised-learning"],"created_at":"2025-05-12T08:06:57.841Z","updated_at":"2026-05-05T21:31:52.620Z","avatar_url":"https://github.com/moses000.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🧠 Noc-netIntel – AI-Powered Network Operations Intelligence Assistant\n\n**Noc-netIntel** is an intelligent chat-driven platform designed to forecast critical network outages, provide possible root cause explanations, suggest proactive FME (Field Maintenance Engineer) deployment, and recommend resolutions — all powered by advanced AI, NLP, and time-series forecasting.\n\n\n## 💡 What It Does\n\n- 🔮 **Predicts outages** today, tomorrow, and over the week\n- 📉 **Identifies root causes** using LSTM + LLM reasoning\n- 📅 **Schedules field engineers** proactively\n- 🧠 **Suggests resolutions** from a growing knowledge base\n- 💬 **Conversational interface** with LLM (DeepSeek/OpenAI-compatible)\n\n\n## 🛠 Technology Stack\n\n| Layer                  | Tech                                |\n|------------------------|-------------------------------------|\n| **AI/NLP**             | DeepSeek / Custom LLM               |\n| **ML/Forecasting**     | PyTorch + Custom LSTM               |\n| **Backend**            | Python (FastAPI preferred)          |\n| **Frontend**           | JavaScript (React recommended)      |\n| **Database**           | PostgreSQL                          |\n| **Data Pipeline**      | Python Scripts / Celery Tasks       |\n| **Deployment**         | Docker \u0026 Docker Compose             |\n| **Scheduler (optional)**| Celery + Redis for task management |\n\n\n## 🧬 End-to-End Workflow\n\n### 1. 🔗 Data Collection\n- Sources: Sensor logs, BTS data, voltage/current levels, historical tickets, alarms\n- Stored in PostgreSQL (structured) and optional object storage (raw logs)\n\n### 2. 🧹 Data Preprocessing\n- Cleansing missing/nulls, noise filtering\n- Timestamp alignment, interpolation\n- Scaling, encoding categorical signals (battery status, alarm type)\n\n### 3. 🔧 Feature Engineering\n- Temporal signals: time of day, day of week, holiday\n- Environmental: power metrics, weather (optional)\n- Historical: frequency of past outages, lag features\n- Rolling stats: moving average, rate of failure\n\n### 4. 📊 ML Forecasting (PyTorch + LSTM)\n- Input: Sequence of multivariate time series\n- Architecture: Multi-head LSTM → Dense heads (classification + regression)\n- Outputs:\n  - Outage probability\n  - Affected region/site\n  - Possible root cause embeddings\n- Metrics: F1, AUC for classification; RMSE for regression\n\n### 5. 🧠 NLP Reasoning Layer (DeepSeek / LLM)\n- Converts ML output into readable advice\n- Enhances with historical patterns and predefined rules\n- Formats chat response: outage + root cause + FME plan + resolution\n\n### 6. 📅 Proactive FME Scheduler\n- Ranks urgency and location clustering\n- Optimizes FME routing using heuristic or ML-based dispatch\n- Integrates with external calendars/ticketing if needed\n\n\n## 💬 Sample Chat Interaction\n\n**User**: \"What outages are expected tomorrow in the North East zone?\"  \n**Noc-netIntel**:\n\n🛑 Predicted 3 possible outages:\n\n* Site BGH-29 (Power drain) – 87% chance\n* Site TMT-02 (Overload) – 72% chance\n* Site JAK-10 (Backhaul degradation) – 55% chance\n\n📌 Root Causes: Battery degradation, high load demand, backhaul link instability\n🛠 Recommended Actions: Pre-deploy backup power units, initiate remote checks\n👷 FME Suggestion: Team Alpha, report at 06:30 AM\n\n\n\n## 🚀 API Overview\n\n- `POST /chat` – Accepts user prompt, returns AI-generated insight\n- `GET /forecast` – Returns raw model prediction\n- `GET /schedule` – Lists recommended FME deployments\n- `GET /logs` – Access recent outage logs (if allowed)\n\n\u003e Full Swagger UI at: `http://localhost:8000/docs`\n\n\n## 🗃 Sample PostgreSQL Schema\n\nCREATE TABLE outage_forecasts (\n  id SERIAL PRIMARY KEY,\n  site_code TEXT,\n  prediction_date TIMESTAMP,\n  outage_probability FLOAT,\n  root_cause TEXT,\n  fme_plan TEXT,\n  resolution TEXT\n);\n\n## 🐳 Setup and Deployment\n\n### ✅ Prerequisites\n\n* Docker \u0026 Docker Compose\n* Python 3.9+\n* Node.js (for frontend)\n\n### 📦 Running Locally\n\nbash\ngit clone https://github.com/moses000/mysoftware-nocNetIntel\ncd noc-netintel\n\n# Run with Docker Compose\ndocker-compose up --build\n\n\u003e Services:\n\u003e\n\u003e * `backend`: FastAPI ML/NLP engine\n\u003e * `frontend`: React chat UI (optional)\n\u003e * `ml_worker`: PyTorch + model runner\n\u003e * `postgres`: SQL data store\n\n\n## 🔐 Auth \u0026 Roles\n\n* JWT-based auth\n* Roles: Admin, Analyst, Engineer\n* Granular data access policies\n\n\n## 📊 Monitoring \u0026 Logging\n\n* Optional: Add Grafana for real-time alert visualization\n* Backend logs all predictions and user queries\n* Alerts for model drift / threshold breaches\n\n## ✍️ Wiki \u0026 Docs\n\n* 📘 `docs/data-pipeline.md`: Ingestion, ETL, transformations\n* 📘 `docs/model.md`: LSTM architecture, training notes\n* 📘 \\`docs\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmoses000%2Fmysoftware-nocnetintel","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmoses000%2Fmysoftware-nocnetintel","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmoses000%2Fmysoftware-nocnetintel/lists"}