Resumen

Process mining is a discipline that deals with discovery, monitoring, and improvement of real business processes using data extracted from each of the process instances [1]. This approach relies on extracting quality knowledge about the process execution from trustworthy inputs. Representational learning techniques can be applied in process mining to abstract valuable information from information systems to execute several process mining tasks such as clustering, process comparison, anomaly detection and predictive and prescriptive process monitoring . With these representational learning techniques, we can have a better handling of complex data with several attributes and many abstraction levels often encountered in business process logs. This Systematic Literature Review aims to gather current and relevant information about embedding methods in process mining and sequence flows and look for potential gaps in new representational learning architectures.