Claims:
We Claim:
1. A method for extracting factoid associated words from natural language sentences, the method comprising:
creating, by a factoid processing device, an input vector comprising a plurality of parameters for each target word in a sentence inputted by a user, wherein the plurality of parameters for each target word comprise a Part of Speech (POS) vector associated with the target word and at least two words preceding the target word, a word embedding for the target word, a word embedding for a head word of the target word in a dependency parse tree of the sentence, a dependency label for the target word, and a semantic role label of the target word;
processing for each target word, by the factoid processing device, the input vector through a trained neural network comprising a bidirectional Long Short Term Memory (LSTM) layer, an LSTM layer, a dense layer, and a Softmax layer, wherein the trained neural network is trained to identify words associated with each of a plurality of factoids from a plurality of sentences;
assigning, by the factoid processing device, one or more factoid tags to each target word in the sentence based on processing of associated input vector through the trained neural network;
extracting, by the factoid processing device, text associated with one or more factoids of the plurality of factoids from the sentence based on the one or more factoid tags associated with each target word in the sentence, wherein the sentence comprises text associated with the one or more factoids; and
providing, by the factoid processing device, a response to the sentence inputted by the user based on the text associated with the one or more factoids, wherein the response comprises at least one of an answer to a user query and an action corresponding to the user query.

2. The method of claim 1, wherein the plurality of factoids comprises at least one of who, what, when, what, why, or how.

3. The method of claim 1 further comprising determining the plurality of parameters for each target word in the sentence inputted by the user.

4. The method of claim 1, wherein the dependency label for the target word indicates relation of the target word with the head word in the sentence.

5. The method of claim 1 further comprising training the neural network to identify one or more factoid tags for words within sentences.

6. The method of claim 5, wherein training the neural network comprises:
annotating one or more factoid tags to each word in a plurality natural language sentences retrieved from a data repository of natural language sentences comprising a plurality of factoid scenarios; and
inputting, iteratively, the annotated one or more factoid tags associated with each word along with the associated word in a plurality of natural language sentences to the neural network for training.

7. The method of claim 1, wherein the one or more factoid tags comprise a begin who tag, an inside who tag, a begin what tag, an inside what tag, a begin where tag, an inside where tag, a begin when tag, an inside when tag, a begin how tag, an inside how tag, a begin why tag, an inside why tag, or others tag.

8. The method of claim 7, wherein a begin tag marks a first word of text associated with a factoid and an inside tag marks subsequent words of the text associated with the factoid.

9. A factoid processing device for extracting factoid associated words from natural language sentences, the factoid processing 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:
create an input vector comprising a plurality of parameters for each target word in a sentence inputted by a user, wherein the plurality of parameters for each target word comprise a Part of Speech (POS) vector associated with the target word and at least two words preceding the target word, a word embedding for the target word, a word embedding for a head word of the target word in a dependency parse tree of the sentence, a dependency label for the target word, and a semantic role label of the target word;
process for each target word, the input vector through a trained neural network comprising a bidirectional Long Short Term Memory (LSTM) layer, an LSTM layer, a dense layer, and a Softmax layer, wherein the trained neural network is trained to identify words associated with each of a plurality of factoids from a plurality of sentences;
assign one or more factoid tags to each target word in the sentence based on processing of associated input vector through the trained neural network;
extract text associated with one or more factoids of the plurality of factoids from the sentence based on the one or more factoid tags associated with each target word in the sentence, wherein the sentence comprises text associated with the one or more factoids; and
provide a response to the sentence inputted by the user based on the text associated with the one or more factoids, wherein the response comprises at least one of an answer to a user query and an action corresponding to the user query.