Generating predictions from a probabilistic process model

1. A computer implemented iterative method for predictive analytics in a semi-structured process comprising:

updating, iteratively, at least one probability of a probabilistic process model for the semi-structured process based on a completed task, wherein updating the at least one probability of the probabilistic process model comprises:

identifying a present state within the semi-structured process;

deriving, from a process policy of the semi-structured process, a set of tasks, of a plurality of tasks associated with the probabilistic process model, that can be performed, at the identified present state, in conforming to the semi-structured process;

establishing a todo list including the set of tasks that can be presently performed;

defining a cost of each of a plurality of tasks of the todo list;

prioritizing the plurality of tasks of the todo list according to the defined costs; and

recommending a next task from the todo list according to the prioritization.

2. The method of claim 1 , further comprising updating the todo list by one of adding a new task to the todo list and removing the completed task from the todo list.

3. The method of claim 1 , further comprising updating the todo list by removing another task from the todo list.

4. The method of claim 1 , wherein updating the at least one probability of the probabilistic process model comprises updating at least one value of document content corresponding to nodes in the probabilistic process model.

5. The computer readable storage medium of claim 1 , further comprising updating a current state to reflect the probabilistic process model in view of the completed task.

6. The computer readable storage medium of claim 1 , wherein defining the cost further comprises: determining a total cost of a policy from a current state including a current cost function and a future cost function; and determining a transition probability between the current state and a future state.

7. A computer readable storage medium embodying instructions executed by a processor for performing an iterative method for predictive analytics in a semi-structured process, the method comprising:

updating, iteratively, at least one probability of a probabilistic process model for the semi-structured process based on a completed task, wherein updating the at least one probability of the probabilistic process model comprises:

receiving the probabilistic process model;

identifying a present state within the semi-structured process;

deriving, from a process policy of the semi-structured process, a set of tasks, of a plurality of tasks associated with the probabilistic process model, that can be performed, at the identified present state, in conforming to the semi-structured process;

establishing a todo list including the set of tasks that can be presently performed;

defining a cost of each of the plurality of tasks of the todo list;

prioritizing the plurality of tasks of the todo list according to the defined costs; and

recommending a next task from the todo list according to the prioritization.

8. The computer readable storage medium of claim 7 , wherein updating the at least one probability of the probabilistic process model comprises updating the todo list including the plurality of tasks.

9. The computer readable storage medium of claim 8 , wherein updating the todo list comprises one of adding a new task to the todo list and removing the completed task from the todo list.

10. The computer readable storage medium of claim 8 , wherein updating the todo list including the plurality of tasks further comprises removing another task from the todo list.

11. The computer readable storage medium of claim 7 , wherein updating the at least one probability of the probabilistic process model comprises updating at least one value of document content corresponding to nodes in the probabilistic process model.

12. The computer readable storage medium of claim 7 , further comprising updating a current state to reflect the probabilistic process model in view of the completed task.

13. The computer readable storage medium of claim 7 , wherein defining the cost further comprises: determining a total cost of a policy from a current state including a current cost function and a future cost function.

14. The computer readable storage medium of claim 13 , further comprising determining a transition probability between the current state and a future state.

15. A non-transitory computer readable storage medium embodying instructions executed by a processor for predictive analytics in a semi-structured process, the method comprising:

updating a probabilistic process model associated with a plurality of tasks of the semi- structured process upon a task of the plurality of tasks being completed, wherein the completed task is removed from the plurality of tasks of the semi-structured process;

identifying a present state within the semi-structured process;

deriving, from a process policy of the semi-structured process, a set of tasks, of the plurality of tasks that can be performed, at the identified present state, in conforming to the semi-structured process;

establishing a todo list including the set of tasks that can be presently performed of remaining tasks of the plurality of tasks;

defining costs of tasks of the todo list based on a prediction of an updated probabilistic process model;

prioritizing that remaining tasks of the todo list according to the defined costs; and

recommending a next task from the remaining tasks of the todo list according to a prioritization.

16. The computer readable storage medium of claim 15 , further comprising iteratively recommending ones of the remaining tasks upon completion of each task.

17. The computer readable storage medium of claim 16 , further comprising removing completed tasks from the plurality of tasks.

18. The computer readable storage medium of claim 15 , wherein defining the cost further comprises: determining a total cost of a policy from a current state including a current cost function and a future cost function.

19. The computer readable storage medium of claim 18 , further comprising determining a transition probability between the current state and a future state.

20. A computer readable storage medium embodying instructions executed by a processor for performing an iterative method for predictive analytics in a semi-structured process, the method comprising:

updating, iteratively, at least one probability of a probabilistic process model for the semi-structured process based on a completed task, wherein updating the at least one probability of the probabilistic process model comprises:

receiving the probabilistic process model;

identifying a present state within the semi-structured process;

deriving, from a process policy of the semi-structured process, a set of tasks, of a plurality of tasks associated with the probabilistic process model, that can be performed, at the identified present state, in conforming to the semi-structured process;

establishing a todo list including the set of tasks that can be presently performed;

defining a cost of each of the plurality of tasks of the todo list;

prioritizing the plurality of tasks of the todo list according to the defined costs;

determining an exception to the todo list; and

recommending a task from a general list according to the exception.

21. The computer readable storage medium of claim 20 , further comprising updating the todo list including the plurality of tasks, wherein updating the todo list comprises one of adding a new task to the todo list and removing the completed task from the todo list.

22. The computer readable storage medium of claim 20 , further comprising updating the todo list including the plurality of tasks, wherein updating the todo list including the plurality of tasks further comprises removing another task from the todo list.

23. The computer readable storage medium of claim 20 , wherein updating the at least one probability of the probabilistic process model comprises updating at least one value of document content corresponding to nodes in the probabilistic process model.

24. The computer readable storage medium of claim 20 , further comprising updating a current state to reflect the probabilistic process model in view of the exception.

25. The computer readable storage medium of claim 20 , wherein defining the cost further comprises: determining a total cost of a policy from a current state including a current cost function and a future cost function; and determining a transition probability between the current state and a future state.