Systems and methods for information retrieval and knowledge discovery

A method for enhanced information retrieval and knowledge discovery from a plurality of repositories, the method comprising:

• receiving, by a processor, a user input from a user via a user interface, wherein the user input comprises a user requirement and a user preference;
• identifying, by the processor, a user intent and a context of the received user input by parsing a plurality of keywords within the received user input;
• extracting, by the processor, multimedia content from a structured data source based on the identified user intent, the context, the user requirement, and the user preference, wherein the structured data source stores the multimedia content retrieved from the plurality of data repositories, wherein the extracting comprises:
• generating, using a Generative Pre-trained Transformer (GPT) model, one or more prompts for retrieving the multimedia content from the structured data source based on the user intent and the context of the received user input;
• fetching, using the generated one or more prompts, the multimedia content by using Application Programming Interfaces (APIs) and structured query language (SQL) queries from the plurality of data repositories, wherein the multimedia content includes information relevant to the user input;
• processing, by the processor, the extracted multimedia content using a generative Artificial Intelligence (AI) model;
• and
• storing the processed multimedia content in the structured data store along with the source information of the extracted multimedia content;
• generating by the processor, at least one response to the received user input based on the processed multimedia content, wherein the at least one response comprises the processed multimedia content with corresponding to one or more of the plurality of data repositories;
• extracting one or more keywords from the at least one response to generate one or more scores with respect to one or more criteria by matching the user input with the extracted one or more keywords;
• determining, based on the generated one or more scores for the one or more criteria, a final score for the at least one response by combining the generated one or more scores for the one or more for the one or more criteria, wherein determining the final score further comprises:
• assigning a weight for each of the one or more criteria based on the plurality of keywords within the received user input, wherein the one or more criteria include user intent alignment, content appropriateness, user preference match, and user profile match;
• and
• generating the one or more scores for the one or more criteria based on the assigned weight;
• outputting, by the processor, based on the final score, the at least one response including source information (URL) of the extracted multimedia content from the one or more of the plurality of data repositories on the user interface of a user device;
• generating one or more suggestive prompts to the user, wherein the one or more suggestive prompts are generated based on one or more of the user intent, the context, the user requirement, the user preference, the at least one response generated to the received user input, and historical user inputs, wherein the one or more suggestive prompts represent the one or more prompts for the user input associated with follow up searches and questions;
• generating, based on the one or more suggestive prompts, one or more responses;
• and
• outputting the one or more suggestive prompts and the corresponding one or more responses on the user interface of the user device, wherein the one or more responses comprise enhanced information including the processed multimedia content with corresponding links to one or more of the plurality of data repositories.

A non-transitory computer readable medium for enhanced information retrieval and knowledge discovery from a plurality of repositories, comprising a processor-executable instructions that cause a processor to:

• receive a user input from a user via a user interface, wherein the user input comprises a user requirement and a user preference;
• identify a user intent and a context of the received user input by parsing a plurality of keywords within the received user input;
• extract multimedia content from a structured data source based on the identified user intent, the context, the user requirement, and the user preference, wherein the structured data source stores the multimedia content retrieved from a plurality of data repositories, wherein the extract multimedia content by:
• generating, using a Generative Pre-trained Transformer (GPT) model, one or more prompts for retrieving the multimedia content from the structured data source based on the user intent and the context of the received user input;
• fetching, using the generated one or more prompts, the multimedia content by using Application Programming Interfaces (APIs) and structured query language (SQL) queries from the plurality of data repositories, wherein the multimedia content includes information relevant to the user input;
• processing the extracted multimedia content using a generative Artificial Intelligence (AI) model;
• and
• storing the processed multimedia content in the structured data store along with the source information of the extracted multimedia content;
• generate at least one response to the received user input based on the processed multimedia content, wherein the at least one response comprises the processed multimedia content with corresponding links to one or more of the plurality of data repositories;
• extract one or more keywords from the at least one response to generate one or more scores with respect to one or more criteria by matching the user input with the extracted one or more keywords;
• determine, based on the generated one or more scores for the one or more criteria, a final score for the at least one response by combining the generated one or more scores for the one or more for the one or more criteria, wherein determining the final score further comprises:
• assigning a weight for each of the one or more criteria based on the plurality of keywords within the received user input, wherein the one or more criteria include user intent alignment, content appropriateness, user preference match, and user profile match;
• and
• generating the one or more scores for the one or more criteria based on the assigned weight;
• output, based on the final score, the at least one response including source information (URL) of the extracted multimedia content from the one or more of the plurality of data repositories on the user interface of a user device;
• generate one or more suggestive prompts to the user, wherein the one or more suggestive prompts are generated based on one or more of the user intent, the context, the user requirement, the user preference, the at least one response generated to the received user input, and historical user inputs, wherein the one or more suggestive prompts represent the one or more prompts for the user input associated with follow up searches and questions;
• generate, based on the one or more suggestive prompts, one or more responses;
• and
• output the one or more suggestive prompts and the corresponding one or more responses on the user interface of the user device, wherein the one or more responses comprise enhanced information including the processed multimedia content with corresponding links to one or more of the plurality of data repositories.