RAG-Based Document Query and Intelligent Assistant System Developed by Dr. Ozan Önal
Developed by Dr. Ozan Önal, this artificial intelligence system is a customizable intelligent assistant that enables users to upload their own documents and ask questions based on their content using RAG architecture, embeddings, vector search, and local large language model integration.
This project was developed by Dr. Ozan Önal with the objective of creating a customizable, document-grounded artificial intelligence assistant capable of operating on user-specific or institutional knowledge sources.
The system is built on a RAG (Retrieval-Augmented Generation) architecture. Documents uploaded by users are processed, divided into meaningful text chunks, and converted into numerical vector representations using embedding models. When a user submits a question, the system identifies the most relevant content through vector-based similarity search and provides the retrieved context to a locally running large language model.
This approach enables the system to generate responses based primarily on the information sources uploaded by the user or institution rather than relying solely on the general knowledge of a language model.
The project includes the following components:
• RAG-based question-answering architecture
• Local LLM integration
• Embedding-based semantic search
• FAISS-based vector indexing
• Document chunking and information retrieval
• User-specific document separation
• Profession- and user-based information management
• Document upload and management infrastructure
• Document-grounded response generation
• Streamlit-based user interface
• Modular Python architecture
The system is not limited to institutional use. The same infrastructure can also be used to create personalized artificial intelligence assistants.
Users can upload their CVs, academic studies, projects, publications, personal notes, professional documents, or other information sources to create an AI assistant specialized in their own knowledge base.
The system can therefore be adapted for use as a personalized AI assistant, academic assistant, professional chatbot, institutional knowledge assistant, regulatory query system, technical document analysis platform, or educational support assistant.
One implementation of the project is a personalized artificial intelligence assistant that answers questions about Dr. Ozan Önal based on previously uploaded information and documents. Visitors can use this assistant to access information about his professional fields, projects, studies, and areas of expertise.
In institutional scenarios, standards, regulations, procedures, technical documentation, reports, training materials, and internal knowledge resources can be integrated into the system and queried using natural language.
The primary objective of the project is to combine large language models with user-specific or institutional knowledge sources in order to create a more controlled, customizable, context-aware, and knowledge-grounded artificial intelligence assistant.
This applied artificial intelligence project was developed by Dr. Ozan Önal within the fields of artificial intelligence, large language models, natural language processing, and Retrieval-Augmented Generation.