Method and system for normalizing automotive data

1. A method of normalizing data and data format of automotive data associated with connected vehicles and obtained from a plurality of sources, the method comprising:

obtaining from a plurality of sources, a plurality of data entries relating to connected vehicles and presented in different data formats;

enabling selection and ordering of a plurality of manipulating modules configured to manipulate data or data format of said data entries, to yield one or more selected and ordered manipulating modules;

executing the selected and ordered manipulating modules, using a computer processor, on said data entries, to yield a plurality of respective data entries that are normalized in accordance with a predefined data format;

applying machine learning algorithms to the plurality of data entries to update the normalization rules by classifying the various data types, formats, names, usage, origin, and content, so that a format diversity is learned and modeled to a model;

using the model to improve the normalization rules based on an actual diversity of the date entries,

wherein the applying of the machine learning algorithms is carried out on sensors records in order to classify and identify metadata and sensor types, to identify sensor type and to define a policy per sensor type, and applying the policy to data records belonging to the same sensor type,

wherein the manipulating modules are software modules comprising instructions in a computer readable medium configured to cause the computer processor to manipulate: a data type, data name, data format, and data content of the data entries, in accordance with normalization rules, and

wherein the plurality of respective data entries that are normalized in accordance with the predefined data format are used as uniform or common data language to support a plurality of use cases for automotive data consumer software applications.

2. The method according to claim 1 , wherein the selection and ordering are carried out by a human user over a user interface.

3. The method according to claim 1 , wherein the selection and ordering are carried out automatically by a computer processor based on the normalization rules.

4. The method according to claim 1 , further comprising applying machine learning to an incoming stream of the plurality of data entries, to update the normalization rules.

5. The method according to claim 1 , wherein the plurality of manipulating modules comprises a name manipulating module configured to rename an attribute of the data entry, based on a policy rule compliant with the normalization rules.

6. The method according to claim 1 , wherein the plurality of manipulating modules comprises a data type manipulation module configured to change the data type of the data entry from a first data type to a second data type.

7. The method according to claim 1 , wherein the plurality of manipulating modules further comprises a unit transformation module configured to transform a unit type of a metric associated with the data entry from a first unit type to a second unit type.

8. The method according to claim 1 , wherein the plurality of manipulating modules further comprises a custom module, configured to manipulate data or data format of the data entry based on user definition.

9. The method according to claim 1 , wherein the plurality of manipulating modules further comprises a data enrichment module configured to enrich a data record of the data entry with predefined values.

10. The method according to claim 1 , wherein the plurality of manipulating modules further comprises a data de-resolution module configured to reduce an accuracy of a data record of the data entry.

11. The method according to claim 1 , further comprising applying data anonymization on at least one part of an attribute of the data entry rendering it unreadable, after the executing of the selected and ordered manipulating modules.

12. The method according to claim 1 , wherein the manipulating of the data entries in accordance with the normalization rules is carried out by executing similar selected and ordered manipulating modules for data entries from similar data sources or from similar sensors.

13. A system for normalizing data and data format of automotive data associated with connected vehicles and obtained from a plurality of sources, the system comprising:

a data collector configured to obtain from a plurality of sources, a plurality of data entries relating to connected vehicles and presented in different data formats;

a normalization module configured to:

enable selection and ordering of a plurality of manipulating modules configured to manipulate data or data format of said data entries, to yield selected and ordered manipulating modules; and

execute the selected and ordered plurality of manipulating modules, using a computer processor, on said data entries, to yield a plurality of respective data entries that are normalized in accordance with a predefined data format; and

a learning module configured to:

apply machine learning algorithms to the plurality of data entries to update the normalization rules by classifying the various data types, formats, names, usage, origin, and content, so that a format diversity is learned and modeled to a model; and

use the model to improve the normalization rules based on an actual diversity of the date entries,

wherein the applying of the machine learning algorithms is carried out on sensors records in order to classify and identify metadata and sensor types, to identify sensor type and to define a policy per sensor type, and applying the policy to data records belonging to the same sensor type,

wherein the plurality of manipulating modules are software modules comprising instructions in a computer readable medium configured to cause the computer processor to manipulate: a data type, data name, data format, and data content of the data entries, in accordance with normalization rules, and

wherein the plurality of respective data entries that are normalized in accordance with the predefined data format are used as uniform or common data language to support a plurality of use cases for automotive data consumer software applications.

14. The system according to claim 13 , wherein the selection and ordering are carried out by a human user over a user interface.

15. The system according to claim 13 , wherein the selection and ordering are carried out automatically by a computer processor based on the normalization rules.

16. The system according to claim 13 , further comprising a learning module configured to apply machine learning to an incoming stream of the plurality of data entries, to update the normalization rules.

17. The system according to claim 13 , wherein the plurality of manipulating modules comprise a name manipulation module configured to rename an attribute of the data entry, based on a policy rule compliant with the normalization rules.

18. The system according to claim 13 , wherein the plurality of manipulating modules comprise a data type manipulation module configured to manipulate the data type of the data entry from a first data type to a second data type.

19. The system according to claim 13 , wherein the plurality of manipulating modules further comprise a unit transformation module configured to transform a unit type of a metric associated with the data entry from a first unit type to a second unit type.

20. A non-transitory computer readable medium for normalizing data and data format of automotive data associated with connected vehicles and obtained from a plurality of sources, the non-transitory computer readable medium comprising a set of instructions that, when executed, cause at least one computer processor to:

obtain from a plurality of sources, a plurality of data entries relating to connected vehicles and presented in different data formats;

enable selection and ordering of a plurality of manipulating modules configured to manipulate data or data format of said data entries, to yield selected and ordered manipulating modules;

execute the selected and ordered manipulating modules, using a computer processor, on said data entries, to yield a plurality of respective data entries that are normalized in accordance with a predefined data format;

apply machine learning algorithms to the plurality of data entries to update the normalization rules by classifying the various data types, formats, names, usage, origin, and content, so that a format diversity is learned and modeled to a model; and

use the model to improve the normalization rules based on an actual diversity of the date entries,

wherein the applying of the machine learning algorithms is carried out on sensors records in order to classify and identify metadata and sensor types, to identify sensor type and to define a policy per sensor type, and applying the policy to data records belonging to the same sensor type,

wherein the plurality of manipulating modules are software modules comprising instructions on the non-transitory computer readable medium configured to cause the computer processor to manipulate: a data type, data name, data format, and data content of the data entries, in accordance with normalization rules, and

wherein the plurality of respective data entries that are normalized in accordance with the predefined data format are used as uniform or common data language to support a plurality of use cases for automotive data consumer software applications.