Method and system for determining an occurrence of a visit to a venue by a user

1. A computer-implemented method of determining an occurrence of a visit to a venue by a user associated with an electronic device, the method being executed by a server, the method comprising, at a training phase:

receiving, by the server, a training SSID;

receiving, by the server, one or more training venue identifiers, each of the one or more training venue identifiers comprising information associated with a respective training venue;

generating, by the server, a positive training pair, the positive training pair including the training SSID paired with a first training venue identifier associated with the training SSID;

generating, by the server, a set of negative training pairs, the set of negative training pairs including at least a second training venue identifier paired with the training SSID, the second training venue identifier not associated with the training SSID; wherein

the positive training pair and the set of negative training pairs form a set of training data;

training a machine learning algorithm (MLA), the training including:

determining, by the server, a set of features representative of a property of the set of training data, the set of features including:

a set of relational features indicative of a relational linkage between the one or more training venue identifiers and the training SSID;

a set of SSID-specific features indicative of one or more characteristics of the training SSID; and

a set of training venue specific features indicative of one or more characteristics of the one or more training venue identifiers; and

generating an inferred function based on the set of features, the inferred function being configured to determine a confidence parameter of a given pair including a given SSID received from the electronic device of the user paired to a given venue identifier, the confidence parameter representing a probability of an occurrence of a visit by the user to the venue associated with the given venue identifier.

2. The method of claim 1 , the method further comprising, at an in-use phase:

receiving, from the electronic device, by the server, a SSID previously accessed by the electronic device;

receiving, by the server, a first venue identifier and a second venue identifier based on the SSID;

generating, by the server, a first pair comprising the SSID and the first venue identifier and a second pair comprising the SSID and the second venue identifier;

determining, using the MLA, the confidence parameter of each of the first pair and second pair based on the set of linking features and the set of non-linking features associated with each of the first pair and the second pair; and

in response to determining that the confidence parameter of the first pair is above the confidence parameter of the second pair, determining, by the server, the occurrence of the visit by the user to a venue associated with the first venue identifier.

3. The method of claim 2 , wherein the first venue identifier being associated with a first venue location and the second venue identifier being associated with a second venue location, the method further comprising:

receiving, from the electronic device, an access location at which the SSID has been accessed by the electronic device; and

receiving the first venue identifier and the second venue identifier in response to the first venue location and the second venue location being within a predetermined distance from the access location.

4. The method of claim 2 , wherein the method further comprises:

receiving an electronic device ID associated with the electronic device;

determining a profile parameter of the user based on the electronic device ID, the profile parameter being indicative of profile characteristics of the user;

in response to the determining the occurrence of the visit by the user to the venue, selecting a digital content item based on the profile parameter, the digital content item being associated with the venue; and

transmitting the digital content item to the electronic device for display.

5. The method of claim 1 , wherein a ratio of the positive training pair to the set of negative training pairs within the set of training data is 1 to 40.

6. The method of claim 1 , wherein the method further comprises:

receiving a training SSID location, the training SSID location corresponding to a geographical location associated with the training SSID;

analyzing a plurality of training venue identifiers, each of the training venue identifier included within the plurality of training venue identifiers being associated with a training venue location; and wherein

receiving the one or more training venue identifiers comprises receiving one or more training venue identifiers, each having a respective training venue location in a predetermined distance from the training SSID location.

7. The method of claim 6 , wherein the predetermined distance corresponds to 75 metres from the training SSID location.

8. The method of claim 6 , wherein the generating the positive training pair comprises:

pairing the training SSID with the first training venue identifier in response to the training SSID location corresponding to the training venue location associated with the first training venue identifier; and wherein

the generating the set of negative training pairs comprises:

pairing the training SSID with at least the second training venue identifier in response to the training SSID location being different to the training venue location associated with the second training venue identifier.

9. The method of claim 1 , wherein for a given training venue identifier, the information associated with a given training venue comprises at least one of:

a name of the training venue;

an operating status of the training venue;

a location of the training venue;

a category associated with the training venue; and

one or more URLs associated with the training venue.

10. The method of claim 9 , wherein the method further comprises:

receiving a training SSID location, the training SSID location corresponding to a geographical location associated with the training SSID; and wherein

the information associated with the given training venue further comprises a training venue location indicative of a location of the training venue; and

the set of relational features comprises at least one of:

a distance between the training SSID location and the training venue location;

a cosine similarity between character frequency vectors of the name of the training venue and the training SSID;

a cosine similarity between trigram frequency vectors of the name of the training venue and the training SSID;

a cosine similarity between token frequency vectors of the name of the training venue and the training SSID; and

a cosine similarity between TF-IDF vectors of the name of the training venue and the training SSID.

11. The method of claim 9 , wherein the set of SSID-specific features comprises at least one of:

a character length of the training SSID; and

a number of lexical tokens within the training SSID.

12. The method of claim 9 , wherein the set of non-linking features comprises at least one of:

the operating status of the training venue;

the category associated with the venue; and

a number of URL variations associated with the training venue.

13. The method of claim 1 , wherein the MLA is using a Friedman's gradient boosting decision trees model.

14. A server for determining an occurrence of a visit to a venue by a user associated with an electronic device, the server comprising a processor configured to, at a training phase:

receive a training SSID;

receive one or more training venue identifiers, each of the one or more training venue identifiers comprising information associated with a respective training venue;

generate a positive training pair, the positive training pair including the training SSID paired with a first training venue identifier associated with the training SSID;

generate a set of negative training pairs, the set of negative training pairs including at least a second training venue identifier paired with the training SSID, the second training venue identifier not associated with the training SSID; wherein

the positive training pair and the set of negative training pairs form a set of training data;

train a machine learning algorithm (MLA), to train the MLA the processor being configured to:

determine a set of features representative of a property of the set of training data, the set of features including:

a set of relational features indicative of a relational linkage between the one or more training venue identifiers and the training SSID;

a set of SSID-specific features indicative of one or more characteristics of the training SSID; and

a set of training venue specific features indicative of one or more characteristics of the one or more training venue identifiers; and

generate an inferred function based on the set of features, the inferred function being configured to determine a confidence parameter of a given pair including a given SSID received from the electronic device of the user paired to a given venue identifier, the confidence parameter representing a probability of an occurrence of a visit by the user to the venue associated with the given venue identifier.

15. The server of claim 14 , the processor is further configured to, at an in-use phase:

receive, from the electronic device, a SSID previously accessed by the electronic device;

receive a first venue identifier and a second venue identifier based on the SSID;

generate a first pair comprising the SSID and the first venue identifier and a second pair comprising the SSID and the second venue identifier;

determine, using the MLA, the confidence parameter of each of the first pair and second pair based on the set of linking features and the set of non-linking features associated with each of the first pair and the second pair; and

in response to determining that the confidence parameter of the first pair is above the confidence parameter of the second pair, determine the occurrence of the visit by the user to a venue associated with the first venue identifier.

16. The server of claim 15 , wherein the first venue identifier is associated with a first venue location and the second venue identifier is associated with a second venue location, the processor being further configured to:

receive, from the electronic device, an access location at which the SSID has been accessed by the electronic device; and

receive the first venue identifier and the second venue identifier in response to the first venue location and the second venue location being within a predetermined distance from the access location.

17. The server of claim 15 , the processor being further configured to:

receive an electronic device ID associated with the electronic device;

determine a profile parameter of the user based on the electronic device ID, the profile parameter being indicative of profile characteristics of the user;

in response to the determining the occurrence of the visit by the user to the venue, select a digital content item based on the profile parameter, the digital content item being associated with the venue; and

transmit the digital content item to the electronic device for display.

18. The server of claim 14 , the processor being further configured to:

receive a training SSID location, the training SSID location corresponding to a geographical location associated with the training SSID;

analyze a plurality of training venue identifiers, each of the training venue identifier included within the plurality of training venue identifiers being associated with a training venue location; and wherein

to receive the one or more training venue identifiers, the processor is configured to receive one or more training venue identifiers each having a respective training venue location in a predetermined distance from the training SSID location.

19. The server of claim 18 , wherein to generate the positive training pair, the processor is configured to:

pair the training SSID with the first training venue identifier in response to the training SSID location corresponding to the training venue location associated with the first training venue identifier; and wherein

to generate the set of negative training pairs, the processor is configured to:

pair the training SSID with at least the second training venue identifier in response to the training SSID location being different to the training venue location associated with the second training venue identifier.

20. The server of claim 14 , wherein for a given training venue identifier, the information associated with a given training venue comprises at least one of:

a name of the training venue;

an operating status of the training venue;

a location of the training venue;

a category associated with the training venue; and

one or more URLs associated with the training venue.