Computer-implemented methods for system identification and connectivity creation. Aspects include obtaining, by the processor, customer data and applying one or more natural language processing techniques to the customer data to generate one or more entities for a first data structure. Aspects also include generating one or more predicted entities based on the one or more entities and analyzing the one or more predicted entities of the first data structure to determine a candidate source system, wherein the one or more predicted entities include one or more connection parameters. Aspects further include generating a system environment to connect the candidate source system to the target system based on the first data structure.