Abstract: The present disclosure relates to the method and system for identifying a task sequence from interaction stream. Method includes receiving interaction stream related to one or more interactions of user with computing system, one or more events that occurred from one or more interactions. The processed interaction stream is transformed into n-grams. Thereafter, a plurality of potential data candidates is identified for each of n-grams by interpreting corresponding start markers and end markers. Thereafter, method includes transforming each of identified plurality of potential data candidates into corresponding potential data candidate vector, and determining similarity score for each pair of plurality of potential data candidates by comparing each of plurality of potential data candidate vectors of corresponding pair of the plurality of potential data candidates.

Abstract: The present disclosure relates to a method for determining optimal process flow (107) for a process performed by users (101) in a computing system. The method comprises receiving plurality of keystroke level events for each of plurality of process flows associated with the process, performed by plurality of users (101). Further, the method comprises determining one or more subtasks (105) for each of the plurality of process flows (102). A subtask comprises a set of events occurring together in the plurality of keystroke level events (103). Furthermore, the method comprises grouping one or more process flows from the plurality of process flows (102) to form one or more clusters (106), based on the one or more subtasks (105). Each cluster comprises the one or more process flows with similar subtasks. Thereafter, the method comprises determining an optimal process flow (107) in each cluster.

Abstract: The present disclosure relates to a method for determining optimal process flow (107) for a process performed by users (101) in a computing system. The method comprises receiving plurality of keystroke level events for each of plurality of process flows associated with the process, performed by plurality of users (101). Further, the method comprises determining one or more subtasks (105) for each of the plurality of process flows (102). A subtask comprises a set of events occurring together in the plurality of keystroke level events (103). Furthermore, the method comprises grouping one or more process flows from the plurality of process flows (102) to form one or more clusters (106), based on the one or more subtasks (105). Each cluster comprises the one or more process flows with similar subtasks. Thereafter, the method comprises determining an optimal process flow (107) in each cluster.