• event log
• process discovery
• process mining
• stigmergy

The thesis work focuses on the development of a temporal approach to process mining, proposing a process discovery algorithm that incorporates the concept of time through the aggregation of tokens via computational stigmergy. More specifically, the algorithm employs what are referred to as Stigmergic Receptive Fields (SRFs). These SRFs can be thought of as computational units, each with the responsibility of creating an aggregated representation of time series data derived from event logs. This approach aims to generate a process map that not only identifies the underlying structure of a process but also highlights the various temporal behaviors associated with it. This temporal approach can be seen as an extension of the Fuzzy Miner, while also taking into consideration the temporal patterns that exist within event logs. This incorporation of temporal patterns adds a layer of complexity to the analysis, potentially revealing insights that were previously hidden. To evaluate the effectiveness of this approach, a series of pilot experiments have been conducted and reported, allowing for empirical validation and the refinement of the algorithm.