Method and system for generating an object card

1. A computer-implemented method for generating an object card, the method being executed by a server connectable to an electronic device associated with a user via a communication network, the server being further coupled to:

a news database comprising a plurality of digital news articles, each of the digital news article having been previously crawled from a respective web resource and comprising an original corpus of text;

an object database, the object database hosting an indication of:

a plurality of objects; and

information data associated with each of the plurality of objects;

the method comprising:

receiving, from the electronic device, a request for a digital news article;

retrieving the digital news article, the digital news article having a respective original corpus of text;

parsing the corpus of text to identify an object contained therein, the object corresponding to one of the plurality of objects;

determining, for the object, a set of object features, the set of object features comprising a set of statistical features being indicative of a statistical characteristic of the object within at least one of:

the digital news article; and

the plurality of digital news articles;

generating, by a machine learning algorithm (MLA), a score value for the object, the score value being representative of a likelihood of the user executing a web search to locate additional information in respect to the object;

in response to the score value being above a predetermined threshold, generating an object card, the object card comprising the object and the associate information data;

transmitting the object card and the digital news article to the electronic device for displaying the object card in association with the digital news article with the original corpus of text.

2. The method of claim 1 , wherein

the MLA is a neural network; and

the method further comprises training the neural network using a training set of data prior to receiving the request for the digital news article.

3. The method of claim 2 , wherein the server is further coupled to:

a browsing log, the browsing log storing an indication of a plurality of navigational session transition patterns, each of the navigational session transition patterns having one or more web pages accessed by a given electronic device; and

the training sets of data comprising at least:

a training set of object features associated with a training object;

a label associated with the training object, the label being indicative of a number of navigational session transition patterns, each of the navigational session transition patterns comprising a first web page associated with a training digital news article comprising the training object, and a second web page associated with a search engine result page comprising the training object as a search term.

4. The method of claim 3 , wherein

the training set of object features comprises a training set of statistical features being indicative of the statistical characteristics of the training object within at least one of the digital news article and the plurality of digital news articles; and

the label comprises a ratio of the number of navigational session transition patterns with regards to the plurality of navigational session transition patterns.

5. The method of claim 3 , wherein the first web page and the second web page have been accessed within a pre-determined time period within each of the number of navigational session transition patterns.

6. The method of claim 5 , wherein the training the neural network comprises:

inputting the training set of object features associated with the training object as an input data;

inputting the label associated with the training object; determining, for the set of object features and the label, a set of features representative of a property of the training object; and

based on the set of features of the training object, learning a training score value, the training score value being indicative of one of a higher degree of likelihood or a lower degree of likelihood of the user executing a web search to locate additional information in respect to the training object after being exposed to the training article.

7. The method of claim 3 , wherein:

the electronic device is associated with a user ID;

each of the number of navigational session transition patterns is associated with the user ID; and

the score value being representative of a respective likelihood of the user associated with the electronic device executing the web search to locate additional information in respect to the object.

8. The method of claim 1 , wherein the set of statistical features comprises at least one of:

a number of occurrence of the indication of the object within the digital news article;

a size of the corpus of text;

an average number of objects present in each of the plurality of digital news articles; and

the set of object features further comprises at least one of:

a set of profile features associated with the object, the profile feature being a set of vectors representing the profile of the text object; and

a popularity feature, the popularity feature being indicative of a popularity of the object as a search term in a search engine service.

9. The method of claim 8 , wherein:

the set of profile features is generated by analyzing a webpage associated with the object; and

the popularity feature is generated by analyzing one of a search history log and a vertical search history log associated with the search engine service.

10. The method of claim 8 , the method further comprising:

parsing the plurality of digital news articles by topic;

identifying a subset of the plurality of digital news articles sharing a same topic with the digital news article; and

wherein the set of object features of the object further comprises at least one of:

a number of digital news articles within the subset comprising the object.

11. The method of claim 1 , wherein the information data is at least one of:

an image;

a text; and

a video.

12. A system for generating an object card, the system comprising a server connectable to:

an electronic device associated with a user via a communication network;

a news database comprising a plurality of digital news articles, each of the digital news article having been previously crawled from a respective web resource and comprising an original corpus of text;

an object database, the object database hosting an indication of:

a plurality of objects; and

information data associated with each of the plurality of objects;

the server comprising a processor configured to:

receive, from the electronic device, a request for a digital news article;

retrieve the digital news article, the digital news article having a respective original corpus of text;

parse the corpus of text to identify an object contained therein, the object corresponding to one of the plurality of objects;

determine, for the object, a set of object features, the set of object features comprising a set of statistical features being indicative of a statistical characteristic of the object within at least one of:

the digital news article; and

the plurality of digital news articles;

generate, by a machine learning algorithm (MLA), a score value for the object, the score value being representative of a likelihood of the user executing a web search to locate additional information in respect to the object;

in response to the score value being above a predetermined threshold, generate an object card, the object card comprising the object and the associate information data;

transmit the object card and the digital news article to the electronic device for displaying the object card in association with the digital news article with the original corpus of text.

13. The system of claim 12 , wherein

the MLA is a neural network; and

the processor is further configured to train the neural network using a training set of data prior to receiving the request for the digital news article.

14. The system of claim 13 , wherein the server is further coupled to:

a browsing log, the browsing log storing an indication of a plurality of navigational session transition patterns, each of the navigational session transition patterns having one or more web pages accessed by a given electronic device; and

the training sets of data comprising at least:

a training set of object features associated with a training object;

a label associated with the training object, the label being indicative of a number of navigational session transition patterns, each of the navigational session transition patterns comprising a first web page associated with a training digital news article comprising the training object, and a second web page associated with a search engine result page comprising the training object as a search term.

15. The system of claim 14 , wherein

the training set of object features comprises a training set of statistical features being indicative of the statistical characteristics of the training object within at least one of the digital news article and the plurality of digital news articles; and

the label comprises a ratio of the number of navigational session transition patterns with regards to the plurality of navigational session transition patterns.

16. The system of claim 14 , wherein the first web page and the second web page have been accessed within a pre-determined time period within each of the number of navigational session transition patterns.

17. The system of claim 16 , wherein to train the neural network, the processor is configured to:

input the training set of object features associated with the training object as an input data;

input the label associated with the training object; determining, for the set of object features and the label, a set of features representative of a property of the training object; and

based on the set of features of the training object, learn a training score value, the training score value being indicative of one of a higher degree of likelihood or a lower degree of likelihood of the user executing a web search to locate additional information in respect to the training object after being exposed to the training article.

18. The system of claim 14 , wherein:

the electronic device is associated with a user ID;

each of the number of navigational session transition patterns is associated with the user ID; and

the score value being representative of a respective likelihood of the user associated with the electronic device executing the web search to locate additional information in respect to the object.

19. The system of claim 12 , wherein the set of statistical features comprises at least one of:

a number of occurrence of the indication of the object within the digital news article;

a size of the corpus of text;

an average number of objects present in each of the plurality of digital news articles; and

the set of object features further comprises at least one of:

a set of profile features associated with the object, the profile feature being a set of vectors representing the profile of the text object; and

a popularity feature, the popularity feature being indicative of a popularity of the object as a search term in a search engine service.

20. The system of claim 18 , wherein:

the set of profile features being generated by analyzing a webpage associated with the object; and

the popularity feature is generated by analyzing one of a search history log and a vertical search history log associated with the search engine service.

21. The system of claim 19 , wherein the processor is further configured to:

parse the plurality of digital news articles by topic;

identify a subset of the plurality of digital news articles sharing a same topic with the digital news article; and

wherein the set of object features of the object further comprises at least one of:

a number of digital news articles within the subset comprising the object.