Method, apparatus, and system for providing a location-aware evaluation of a machine learning model

1. A computer-implemented method for providing a location-aware evaluation of a machine learning model comprising:

designating a geographic area for creating an evaluation dataset for the machine learning model; and

separating, by a processor, a plurality of observation data records into the evaluation dataset and a training dataset based on a comparison of a respective data collection location of each of the plurality of observation data records to the geographic area,

wherein the training dataset is used to train the machine learning model; and

wherein the evaluation dataset is used to evaluate the trained machine learning model.

2. The method of claim 1 , wherein the plurality of observation data records is captured by one or more sensors of one or more probe devices.

3. The method of claim 1 , wherein the plurality of observation data records includes one or more location-tagged images.

4. The method of claim 1 , further comprising:

adding at least one record of the plurality of observation data records to the evaluation dataset based on determining that the respective data collection location is within the geographic area.

5. The method of claim 1 , further comprising:

adding at least one record of the plurality of observation data records to the training dataset based on determining that the respective data collection location is not within the geographic area.

6. The method of claim 1 , further comprising:

receiving another observation data record after creating the evaluation dataset, the training dataset, or a combination thereof; and

expanding the evaluation dataset or the training dataset to include the another observation data record based on a comparison of another data collection location of the another observation data record to the geographic area.

7. The method of claim 1 , wherein the geographic area is specified as a geo-fence.

8. The method of claim 1 , wherein the geographic area is designated based on one or more map attributes.

9. The method of claim 8 , wherein the geographic area is selected from one or more candidate geographic areas based on a diversity of the one or more map attributes.

10. The method of claim 1 , wherein the one or more attributes include a functional class, a road elevation, a speed category, a presence or absence of road features, or a combination thereof.

11. An apparatus for providing a location-aware evaluation of a machine learning model comprising:

at least one processor; and

at least one memory including computer program code for one or more programs,

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following,

designating a geographic area for creating a training dataset for the machine learning model; and

separating a plurality of observation data records into the training dataset and an evaluation dataset based on a comparison of a respective data collection location of each of the plurality of observation data records to the geographic area,

wherein the training dataset is used to train the machine learning model; and

wherein the evaluation dataset is used to evaluate the trained machine learning model.

12. The apparatus of claim 11 , wherein the apparatus is further caused to:

add at least one record of the plurality of observation data records to the training dataset based on determining that the respective data collection location is within the geographic area.

13. The apparatus of claim 11 , wherein the apparatus is further caused to:

add at least one record of the plurality of observation data records to the evaluation dataset based on determining that the respective data collection location is not within the geographic area.

14. The apparatus of claim 11 , wherein the apparatus is further caused to:

receive another observation data record after creating the training dataset, the evaluation dataset, or a combination thereof; and

expand the training dataset or the evaluation dataset to include the another observation data record based on a comparison of another data collection location of the another observation data record to the geographic area.

15. The apparatus of claim 11 , wherein the geographic area is designated based on one or more map attributes.

16. A non-transitory computer-readable storage medium for providing a location-aware evaluation of a machine learning model, carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to perform:

designating a geographic area for creating an evaluation dataset for the machine learning model; and

separating a plurality of observation data records into the evaluation dataset and a training dataset based on a comparison of a respective data collection location of each of the plurality of observation data records to the geographic area,

wherein the training dataset is used to train the machine learning model; and

wherein the evaluation dataset is used to evaluate the trained machine learning model.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the apparatus is further caused to perform:

adding at least one record of the plurality of observation data records to the evaluation dataset based on determining that the respective data collection location is within the geographic area.

18. The non-transitory computer-readable storage medium of claim 16 , wherein the apparatus is further caused to perform:

adding at least one record of the plurality of observation data records to the training dataset based on determining that the respective data collection location is not within the geographic area.

19. The non-transitory computer-readable storage medium of claim 16 , wherein the apparatus is further caused to perform:

receiving another observation data record after creating the training dataset, the evaluation dataset, or a combination thereof; and

expanding the training dataset or the evaluation dataset to include the another observation data record based on a comparison of another data collection location of the another observation data record to the geographic area.

20. The non-transitory computer-readable storage medium of claim 16 , wherein the geographic area is designated based on one or more map attributes.