Data-driven execution system and execution method thereof

1. A data-driven execution system, comprising:

a storage device, used to store a plurality of modules; and

a processor, coupled to the storage device and used to execute the modules,

wherein the processor receives a data processing request, and the processor searches a data graph according to the data processing request to obtain at least one task path,

wherein the processor generates a recommended plan according to one of the at least one task path,

wherein the processor determines a matching degree of a data feature and the at least one task path based on a current data status of the data processing request to select a task path with a highest matching degree, and the processor generates the recommended plan according to the task path with the highest matching degree,

wherein the processor separately calculates a plurality of data conditions of each of the at least one task path to obtain a plurality of matching scores, and sums up the matching scores to obtain the matching degree of each of the at least one task path.

2. The data-driven execution system according to claim 1 , wherein the at least one task path is a task list built from a plurality of data models, and each of the data models is formed by corresponding input data, a corresponding task, and corresponding output data.

3. The data-driven execution system according to claim 1 , wherein the processor performs data packaging on initial data of the data processing request to generate a data instance; and the processor searches the data graph according to the data instance and a target to realize of the data processing request to obtain the at least one task path.

4. The data-driven execution system according to claim 1 , wherein the processor selects one of a plurality of task paths according to a preset rule to generate the recommended plan when the plurality of task paths have matching degrees that are the same.

5. The data-driven execution system according to claim 1 , wherein the processor generates a real-time model according to a path selection command provided by a user interface, and the processor generates the recommended plan according to the real-time model and a pre-stored offline model.

6. The data-driven execution system according to claim 1 , wherein the one of the at least one task paths used to generate the recommended plan has a shortest execution path, a shortest execution time, or a highest recommendation degree.

7. The data-driven execution system according to claim 1 , wherein the processor predetermines a plurality of reference data and a plurality of reference statuses, and predetermines a plurality of reference targets and creates a plurality of reference tasks, wherein the processor creates a plurality of reference plans according to the reference data, the reference statuses, the reference targets, and the reference tasks, wherein the processor selects one of the reference plans for generating the recommended plan.

8. The data-driven execution system according to claim 7 , wherein the processor creates the data graph in advance based on a relationship between the reference data and the reference tasks, and stores the data graph into the storage device.

9. A data-driven execution method, comprising:

receiving a data processing request;

searching a data graph according to the data processing request to obtain at least one task path; and

generating a recommended plan according to one of the at least one task path,

wherein generating the recommended plan comprises:

determining a matching degree of a data feature and the at least one task path based on a current data status of the data processing request to select a task path with a highest matching degree; and

generating the recommended plan according to the task path with the highest matching degree, wherein generating the recommended plan further comprises:

separately calculating a plurality of data conditions of each of the at least one task path to obtain a plurality of matching scores, and summing up the matching scores to obtain the matching degree of each of the at least one task path.

10. The data-driven execution method according to claim 9 , wherein generating the recommended plan further comprises:

selecting one of a plurality of task paths according to a preset rule to generate the recommended plan when the plurality of task paths have matching degrees that are the same.

11. The data-driven execution method according to claim 9 , wherein the at least one task path is a task list built from a plurality of data models, and each of the data models is formed by corresponding input data, a corresponding task, and corresponding output data.

12. The data-driven execution method according to claim 9 , wherein obtaining the at least one task path comprises:

performing data packaging on initial data of the data processing request to generate a data instance; and

searching the data graph according to the data instance and a target to realize of the data processing request to obtain the at least one task path.

13. The data-driven execution method according to claim 9 , wherein generating the recommended plan comprises:

generating a real-time model according to a path selection command provided by a user interface; and

generating the recommended plan according to the real-time model and a pre-stored offline model.

14. The data-driven execution method according to claim 9 , wherein the one of the at least one task paths used to generate the recommended plan has a shortest execution path, a shortest execution time, or a highest recommendation degree.

15. The data-driven execution method according to claim 9 , further comprising:

predetermining a plurality of reference data and a plurality of reference statuses;

predetermines a plurality of reference targets and creating a plurality of reference tasks;

creating a plurality of reference plans according to the reference data, the reference statuses, the reference targets, and the reference tasks; and

selecting one of the reference plans for generating the recommended plan.

16. The data-driven execution method according to claim 15 , further comprising:

creating the data graph in advance based on a relationship between the reference data and the reference tasks; and

storing the data graph into the storage device.