Contextual recommendation

1. An electronic device, comprising:

a memory storing instructions; and

at least one processor configured to execute the instructions to:

receive a user input message at the electronic device to communicate with another electronic device,

receive from the other electronic device a message including a word,

display the message on a display of the electronic device,

identify, based on a language model, an application corresponding to the message and the user input message,

identify, in the user input message, a reference object indicated by the word of the message based on the identified application,

highlight the word in the displayed message,

access a local database to obtain a user in put history including the user input message and represent the user input history in a form of a feature vector of floating point numbers, and

display, based on receiving a selection of the highlighted word in the displayed message, recommendations for completing an input to reply to the message, including the identified reference object,

wherein the language model is trained based on the displaying of the recommendations in the identified application, the identified reference object indicated by the word, and the identified application.

2. The electronic device of claim 1 ,

wherein the identified application is executable by the at least one processor, and

wherein the displaying of the recommendations is a function of the identified application.

3. The electronic device of claim 1 , wherein the identified application is correlated mostly with the word among a plurality of applications.

4. The electronic device of claim 3 , wherein a correlation between the word and the identified application is learned by the language model.

5. The electronic device of claim 1 ,

wherein the identified application is running on a foreground of the electronic device,

wherein the message is displayed on the identified application, and

wherein the displaying of the recommendations is a function of the identified application.

6. The electronic device of claim 1 ,

wherein the reference object relates to a hyperlink, and

wherein the reply to the message comprises transmission of the hyperlink.

7. The electronic device of claim 1 ,

wherein the reference object relates to a file, and

wherein the reply to the message comprises attachment of the identified reference object.

8. The electronic device of claim 1 ,

wherein the word comprises a noun word, and

wherein the at least one processor is further configured to execute the instructions to identify a co-reference word of the noun word, the co-reference word indicating the reference object.

9. The electronic device of claim 1 ,

wherein the word comprises an anaphora word, and

wherein the at least one processor is further configured to execute the instructions to identify an antecedent word of the anaphora word, the antecedent word indicating the reference object.

10. The electronic device of claim 1 ,

wherein the word comprises a verb word, and

wherein the at least one processor is further configured to execute the instructions to identify an object word for the verb word, the object word indicating the reference object.

11. The electronic device of claim 1 , wherein the at least one processor is further configured to execute the instructions to use a virtual assistant to reply to the message.

12. The electronic device of claim 1 , wherein the at least one processor is further configured to execute the instructions to use a voice command to reply to the message.

13. A method comprising:

receive a user input message at an electronic device to communicate with another electronic device;

receive from the other electronic device a message including a word;

displaying the received message;

identifying, based on a language model, an application corresponding to the message and the user input message;

identifying, in the user input message, a reference object indicated by the word of the message based on the identified application;

highlighting the word in the displayed message;

accessing a local database to obtain a user input history including the user input message and represent the user input history in a form of a feature vector of floating point numbers; and

displaying, based on receiving a selection of the highlighted word in the displayed message, recommendations for completing an input to reply to the message, including the identified reference object,

wherein the language model is trained based on the displaying of the recommendations in the identified application, the identified reference object indicated by the word, and the identified application.

14. A non-transitory computer readable medium comprising instructions which, when executed by at least one processor of an electronic device, cause the electronic device to:

receive a user input message at the electronic device to communicate with another electronic device,

receive from the other electronic device a message including a word,

display the received message on a display of the electronic device,

identify, based on a language model, an application corresponding to the message and the user input message,

identify, in the user input message, a reference object indicated by the word of the message based on the identified application,

highlight the word in the displayed message,

access a local database to obtain a user input history including the user input message and represent the user input history in a form of a feature vector of floating point numbers, and

display, based on receiving a selection of the highlighted word, recommendations for completing an input to reply to the message, including the identified reference object,

wherein the language model is trained based on the displaying of the recommendations in the identified application, the identified reference object indicated by the word, and the identified application.