Víceúrovňová optimalizace konverzačního asistenta s využitím platformy Cognigy

The thesis deals with the complex optimization of a conversational assistant using the Cognigy platform. The aim is to improve the quality of user interactions and increase the efficiency of process management. Based on the identification of the main areas for improvement, a new orchestration flow is proposed that improves the recognition of user intent and enables more efficient management of chatbot interactions. The system is the basis for future optimizations. The implementation of advanced NLP techniques, including DDS and RAG, and the integration of Language Learning Models (LLM) help to mine information and formulate responses more accurately. The proposed approach brings a higher degree of customization to user queries, greater variability in answers, and the ability to respond effectively to complex questions. The contribution of this work is to contribute to the development of technologies for conversational assistants and to open new avenues for future research in artificial intelligence and automated communication.