The article "Discovering Models of Software Processes from Event-Based Data" by Jonathan E. Cook and Alexander L. Wolf explores methods for automatically deriving formal models of software processes from event-based data. The authors address the challenge of developing formal models for complex, ongoing processes, which can be difficult and costly. They introduce a technique called *process discovery*, which captures process events and generates a formal model of the process behavior. The article presents three methods for process discovery: a Markov method, an RNet method, and a Ktail method. These methods range from purely algorithmic to purely statistical approaches. The Markov method uses Markov models to find the most probable event sequences and converts them into states and transitions. The RNet method is a neural network approach that looks at past behavior to characterize a state. The Ktail method is an algorithmic approach that looks at future behavior to compute a possible current state. The article also discusses the application of these methods in an industrial case study and compares their performance. The goal is to provide process engineers with initial models that can be refined, supporting the improvement and evolution of software processes.The article "Discovering Models of Software Processes from Event-Based Data" by Jonathan E. Cook and Alexander L. Wolf explores methods for automatically deriving formal models of software processes from event-based data. The authors address the challenge of developing formal models for complex, ongoing processes, which can be difficult and costly. They introduce a technique called *process discovery*, which captures process events and generates a formal model of the process behavior. The article presents three methods for process discovery: a Markov method, an RNet method, and a Ktail method. These methods range from purely algorithmic to purely statistical approaches. The Markov method uses Markov models to find the most probable event sequences and converts them into states and transitions. The RNet method is a neural network approach that looks at past behavior to characterize a state. The Ktail method is an algorithmic approach that looks at future behavior to compute a possible current state. The article also discusses the application of these methods in an industrial case study and compares their performance. The goal is to provide process engineers with initial models that can be refined, supporting the improvement and evolution of software processes.