{"id":29178348,"url":"https://github.com/yashdew3/agentic-ai-chat-analyzer","last_synced_at":"2026-04-16T19:03:09.135Z","repository":{"id":301767361,"uuid":"1010189756","full_name":"yashdew3/Agentic-AI-Chat-Analyzer","owner":"yashdew3","description":"💬 Analyze agent conversations using LLMs to generate smart summaries, sentiment insights, and article links — powered by FastAPI + Streamlit + HuggingFace 🚀","archived":false,"fork":false,"pushed_at":"2025-06-28T17:39:35.000Z","size":8,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-06-28T18:33:03.820Z","etag":null,"topics":["agentic-ai","chat-analysis","data-pipeline","docker","fastapi","huggingface","huggingface-transformers","llm","nlp","portfolio-website","sentiment-analysis","streamlit","summarization"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/yashdew3.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":"2025-06-28T14:44:05.000Z","updated_at":"2025-06-28T17:39:38.000Z","dependencies_parsed_at":"2025-06-28T18:33:59.512Z","dependency_job_id":"35199136-800e-40e7-8e47-edf5b022c1e8","html_url":"https://github.com/yashdew3/Agentic-AI-Chat-Analyzer","commit_stats":null,"previous_names":["yashdew3/agentic-ai-chat-analyzer"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/yashdew3/Agentic-AI-Chat-Analyzer","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yashdew3%2FAgentic-AI-Chat-Analyzer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yashdew3%2FAgentic-AI-Chat-Analyzer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yashdew3%2FAgentic-AI-Chat-Analyzer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yashdew3%2FAgentic-AI-Chat-Analyzer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/yashdew3","download_url":"https://codeload.github.com/yashdew3/Agentic-AI-Chat-Analyzer/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yashdew3%2FAgentic-AI-Chat-Analyzer/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":268053868,"owners_count":24188107,"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","status":"online","status_checked_at":"2025-07-31T02:00:08.723Z","response_time":66,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["agentic-ai","chat-analysis","data-pipeline","docker","fastapi","huggingface","huggingface-transformers","llm","nlp","portfolio-website","sentiment-analysis","streamlit","summarization"],"created_at":"2025-07-01T18:46:07.309Z","updated_at":"2026-04-16T19:03:09.083Z","avatar_url":"https://github.com/yashdew3.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🧠 Agentic AI Chat Analyzer\n\n\u003e An AI-powered platform for analyzing agent chat transcripts about Washington Post articles using lightweight LLMs, visual EDA, and an interactive frontend.  \n\u003e Designed to summarize conversations, detect sentiments, and deliver insights — all through a modular, API-driven architecture.\n\n---\n\n## 🚀✨ Features\n✅ Modular Data Pipeline – Clean ingestion, cleaning, and transformation\n\n📊 EDA + Profiling – Dataset summaries, word clouds, sentiment plots\n\n🧠 LLM Summarizer – Uses Flan-T5-small\n\n💬 Sentiment Classification – CardiffNLP RoBERTa Sentiment\n\n🌐 FastAPI Backend – 3 REST endpoints with Pydantic validation\n\n🖼️ Streamlit Frontend – Interactive UI for real-time transcript analysis\n\n📦 Model Caching – Offline-ready with locally saved models\n\n🧪 Evaluation Pipeline – Accuracy and BLEU score metrics\n\n---\n\n## 🗂️ Folder Structure\n```bash\nAgentic_AI_ChatAnalyzer/\n├── app/                # Core app logic (API, services, utils)\n├── data/               # Dataset and output CSVs\n├── models/             # Locally saved HuggingFace models\n├── notebooks/          # EDA and profiling notebook\n├── scripts/            # Model downloader and batch inference\n├── streamlit_app/      # UI client (integrated with FastAPI)\n├── main.py             # FastAPI entrypoint\n├── requirements.txt    # Python dependencies\n├── .gitignore\n└── README.md\n```\n\n---\n\n## 📐 System Architecture\n\n```mermaid\nflowchart LR\n    A[\"User Input via Streamlit UI\"] --\u003e B[\"POST Transcript to FastAPI API\"]\n    B --\u003e C[\"LLM Summarizer (Flan-T5)\"]\n    B --\u003e D[\"Sentiment Classifier (RoBERTa)\"]\n    C --\u003e E[\"Summary Output\"]\n    D --\u003e F[\"Sentiment per Agent\"]\n    E --\u003e G[\"Response JSON to Streamlit\"]\n    F --\u003e G\n    G --\u003e H[\"Visualization: WordClouds + Metrics\"]\n```\n\n\n---\n\n## ⚙️ Setup Instructions\n\n```bash\n# 1. Clone the repo\ngit clone https://github.com/yashdew3/Agentic-AI-Chat-Analyzer.git\ncd Agentic_AI_ChatAnalyzer\n\n# 2. Create and activate environment (optional)\npython -m venv venv\nsource venv/bin/activate   # Windows: venv\\Scripts\\activate\n\n# 3. Install dependencies\npip install -r requirements.txt\n\n# 4. Download local LLM + sentiment models (run once)\npython scripts/download_models.py\n```\n---\n\n## 🚀 How to Use\n### Run FastAPI Backend\n```bash\nuvicorn main:app --reload\n```\n- Access Swagger Docs: http://localhost:8000/docs\n\n---\n### Run Streamlit Frontend\n```bash\nstreamlit run streamlit_app/app_ui.py\n```\n- Analyze transcripts visually\n\n- Paste chats and see live: \n    - `Summary`\n    - `Sentiment Insights`\n    - `Word Clouds`\n    - `Article Links`\n---\n\n## 📡 API Endpoints\n\n| Method | Endpoint     | Description                                        |\n| ------ | ------------ | -------------------------------------------------- |\n| GET    | `/summary`   | Returns dataset stats                              |\n| POST   | `/transform` | Preprocess and clean new chat input                |\n| POST   | `/insights`  | Summarizes, classifies sentiment, returns insights |\n\n---\n\n## 🧪 Sample API Input\n\n### 📥 Sample Payload (for `/insights`)\n```json\n{\n  \"content\": [\n    {\"agent\": \"agent_1\", \"message\": \"Let’s discuss the article on the new football rule change.\"},\n    {\"agent\": \"agent_2\", \"message\": \"Yes, it's causing a lot of debate on ESPN.\"}\n  ],\n  \"article_url\": \"https://www.washingtonpost.com/sports/football-rule-change\"\n}\n\n```\n---\n\n## 📈 Tech Stack\n\n| Layer       | Tools Used                                 |\n| ----------- | ------------------------------------------ |\n| Backend     | FastAPI, Uvicorn, Pydantic                 |\n| Frontend    | Streamlit, Matplotlib, WordCloud           |\n| NLP Models  | HuggingFace Transformers: Flan-T5, RoBERTa |\n| Analysis    | Pandas, Seaborn, ydata-profiling           |\n| LLM Hosting | Locally saved models in `/models/`         |\n---\n\n## 🔮 Future Enhancements\n- 🔐 Add JWT authentication\n\n- 🐳 Dockerize backend \u0026 frontend for deployment\n\n- ☁️ Deploy on Render, Railway, or HuggingFace Spaces\n\n- 📤 Export chat reports (PDF/CSV)\n\n- 💬 Multi-transcript support with database\n\n---\n\n## 📄 License\nMIT License © Yash Dewangan\n\n## ⭐ Show Your Support\nIf you liked this project, give it a ⭐ star on GitHub and feel free to fork it!\n\n## 💬 Let's Connect\nFeel free to connect or suggest improvements!\n- Built by **Yash Dewangan**\n- 🐙Github: [YashDewangan](https://github.com/yashdew3)\n- 📧Email: [yashdew06@gmail.com](mailto:yashdew06@gmail.com)\n- 🔗Linkedin: [YashDewangan](https://www.linkedin.com/in/yash-dewangan/)\n\n## 🤝 Contributing\nContributions, issues, and feature requests are welcome! Feel free to check the [issues page](https://github.com/yashdew3/Agentic-AI-Chat-Analyzer/issues) (if you have one) or open a new issue to discuss changes. Pull requests are also appreciated.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fyashdew3%2Fagentic-ai-chat-analyzer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fyashdew3%2Fagentic-ai-chat-analyzer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fyashdew3%2Fagentic-ai-chat-analyzer/lists"}