Method and device for determining a relation between two or more entities

1. A method of determining a relation between two or more entities in a text document, the method comprising:

receiving, by a relation determining device, training text data annotated with two or more entities;

creating, by the relation determining device, one or more n-grams based on the training text data;

generating, by the relation determining device, a Convolutional Neural Network (CNN) model using the one or more n-grams;

creating, by the relation determining device, an entity vector using at least one of a word embedding and a numeric embedding based on the training text data; and

generating, by the relation determining device, a relation-entity model using the CNN model and the entity vector, wherein generating the relation-entity model comprises:

creating a first set of features using one or more consecutive words of the training text data, based on convolution filter of the CNN model;

generating a second set of features based on one or more fully connected (FC) layers of the CNN model; and

generating the relation-entity model using the CNN model, and the first set of features and the second set of features.

2. The method of claim 1 , wherein the training text data comprises one or more textual sentences.

3. The method of claim 1 , wherein generating the CNN model further comprises:

receiving a base model trained on a corpus of data similar to the training text data;

extracting one or more parameters from the base model; and

generating a CNN model using the one or more parameters wherein at least one of a plurality of layers of the CNN model are trainable using training text data.

4. The method of claim 3 , wherein the one or more parameters comprise one or more learned weights associated with the CNN model.

5. The method of claim 1 , further comprising:

receiving an input text data; and

determining one or more entities and one or more relation between two or more entities from the input text data, using the relation-entity model.

6. The method of claim 5 , further comprising:

receiving from a user a selection of an entity from the input text;

determining one or more relations of the selected entity with one or more entities in the input text;

generating a graphical representation of the relation of the selected entity with the one or more entities in the input text; and

rendering the graphical representation via a graphical user interface (GUI).

7. The method of claim 6 , further comprising:

receiving from a user a feedback on the extracted one or more entities and the one or more relation between the two or more entities; and

updating the relation-entity model based on the feedback.

8. A system for determining a relation between two or more entities in a text document, the system comprising:

a relation determining device comprising a processor and a memory communicatively coupled to the processor, wherein the memory stores processor instructions, which, on execution, causes the processor to:

receive training text data annotated with two or more entities;

create one or more n-grams based on the training text data;

generate a Convolutional Neural Network (CNN) model using the one or more n-grams;

create an entity vector using at least one of a word embedding and a numeric embedding based on the training text data; and

generate a relation-entity model using the CNN model and the entity vector, wherein generating the relation-entity model comprises:

creating a first set of features using one or more consecutive words of the training text data, based on convolution filter of the CNN model;

generating a second set of features based on one or more fully connected (FC) layers of the CNN model; and

generating the relation-entity model using the CNN model, and the first set of features and the second set of features.

9. The system of claim 8 , wherein the training text data comprises one or more textual sentences.

10. The system of claim 8 , wherein generating the CNN model further comprises:

receiving a base model trained on a corpus of data similar to the training text data;

extracting one or more parameters from the base model; and

generating a CNN model using the one or more parameters wherein at least one of a plurality of layers of the CNN model are trainable using training text data.

11. The system of claim 10 , wherein the one or more parameters comprise one or more learned weights associated with the CNN model.

12. The system of claim 8 , wherein the processor instructions further cause the processor to:

receive an input text data; and

determine one or more entities and one or more relation between two or more entities from the input text data, using the relation-entity model.

13. The system of claim 12 , wherein the processor instructions further cause the processor to:

receive from a user a selection of an entity from the input text;

determine one or more relations of the selected entity with one or more entities in the input text;

generate a graphical representation of the relation of the selected entity with the one or more entities in the input text; and

render the graphical representation via a graphical user interface (GUI).

14. The system of claim 13 , wherein the processor instructions further cause the processor to:

receive from a user a feedback on the extracted one or more entities and the one or more relation between the two or more entities; and

update the relation-entity model based on the feedback.

15. A non-transitory computer-readable storage medium having stored thereon, a set of computer-executable instructions causing a computer comprising one or more processors to perform steps comprising:

receiving training text data annotated with two or more entities;

creating one or more n-grams based on the training text data;

generating a Convolutional Neural Network (CNN) model using the one or more n-grams;

creating an entity vector using at least one of a word embedding and a numeric embedding based on the training text data; and

generating a relation-entity model using the CNN model and the entity vector, wherein generating the relation-entity model comprises:

creating a first set of features using one or more consecutive words of the training text data, based on convolution filter of the CNN model;

generating a second set of features based on one or more fully connected (FC) layers of the CNN model; and

generating the relation-entity model using the CNN model, and the first set of features and the second set of features.