Processing event instance data in a client-server architecture

摘要

A process analysis system processes event data describing real-world processes. The process analysis system performs the following acts: importing event instance data sets from an information management system, each set having one or more attributes describing an event instance in the real-world process; for each event instance, determining a corresponding process instance based on at least the attributes; determining event order attribute(s) for each imported event instance data set based on other event instance data sets corresponding to the same process instance; forming an analysis result set based on at least the event instance data sets and at least one first or second attribute; the client(s) presenting an analysis utilizing the analysis result set.

主权项

1. A method for analyzing information derived from event data by a computer-implemented analysis system, which comprises a server and one or more clients, the method comprising the following acts performed by the server:
importing event instance data comprising a plurality of event instance data sets from at least one information management system, wherein each event instance data set comprises one or more attributes describing an event instance in the real-world process, wherein the event data describes a real-world process the execution of which is supported by the at least one information management system but the real-world process is not directly connectable with the computer-implemented analysis system; determining for each imported event instance data set a corresponding process instance based on at least the attributes of the imported event instance data set; determining at least one event order attribute for each imported event instance data set based on other event instance data sets corresponding to the same process instance; calculating order information for each event instance data set based on at least the event order attribute and process instance; determining at least one event type attribute for each imported event instance data set based on at least the attributes of event instance data sets, wherein the determined event type attribute indicates an event type for the event instance; whereby the process instances can be determined and the order information can be calculated regardless of whether a schema describing the real-word process is available; creating a causal model based on at least the event instance data sets; using the causal model to calculate a probability for one or more predicted future event instances for at least one process instance based on at least the causal model; forming an analysis result set based on at least some of the event instance data sets and at least one of the one or more predicted future event instances; sending the analysis result set to one or more clients; wherein the method further comprises: at the one or more clients, presenting an analysis utilizing the analysis result set.