1. A method for user-intent based product recommendations, the method comprising:
receiving, by a product recommendation device, one or more multi-modal user inputs
associated with a first product, wherein the one or more multi-modal user inputs are at least speech, text, bodily expressions and clickstream data;
determining, by the product recommendation device, an intent-score for each of the one or more multi-modal user inputs using one or more trained data models;
computing, by the product recommendation device, an emotion-score by aggregating the intent-score for each of the one or more multi-modal user inputs based on a weighted average of the intent-score for each of the one or more multi-modal user inputs; and
recommending, by the product recommendation device, one or more second products based on the emotion-score.
2. The method of claim 1, wherein the one or more trained data models comprises a natural language processing model, a text model, and a convolution neural network model.
3. The method of claim 2, wherein the natural language processing model, the speech to text model, and the convolution neural network model determines the intent-score for text, the intent-score for speech and the intent-score for bodily expressions respectively.
4. The method as claimed in claim 1, wherein the bodily expressions comprise at least one of hand gestures, eye movements, facial expressions, and head rotation, and age.
5. The method of claim 1, wherein the intent-score associated with the one or more multi-modal inputs is further adjusted dynamically based on the clickstream data of the user.
6. The method of claim 5, wherein the intent-scores adjusted are determined into a preconfigured range based on the clickstream data of the user.
7. A product recommendation device for recommending products to a user comprising:
a processor; and
a memory communicatively coupled to the processor, wherein the memory stores processor executable instructions, which on execution causes the processor to:
receive one or more multi-modal user inputs associated with a first product, wherein the one or more multi-modal user inputs are at least speech, text, bodily expressions and clickstream data;

determine an intent-score for each of the one or more multi-modal user inputs using one or more trained data models;
compute an emotion-score by aggregating the intent-score for each of the one or more multi-modal user inputs based on a weighted average of the intent-score for each of the one or more multi-modal user inputs;
recommending, by the product recommendation device, one or more second products based on the emotion-score.
8. The product recommendation device of claim 7, wherein the one or more trained data models comprises a natural language processing model, a speech to text model, and a convolution neural network model.
9. The product recommendation device of claim 8, wherein the natural language processing model, the speech to text model, and the convolution neural network model determines the intent-score for text, the intent-score for speech to text and the intent-score for bodily expressions respectively.
10. The product recommendation device of claim 7, wherein the bodily expressions comprise at least one of hand gestures, eye movements, facial expressions, and head rotation, and age.
11. The product recommendation device of claim 7, wherein the intent-score associated with the one or more multi-modal inputs is further adjusted dynamically based on the clickstream data of the user.
12. The product recommendation device of claim 11, wherein the intent-scores adjusted are determined into a preconfigured range based on the clickstream data of the user.