An Improved Information Retrieval System using Hybrid RNN LSTM for Multiple Search Engines

When searching for data on the internet, every user has their own personal context for doing so. The job of a search engine is to find the most relevant content from all the blogs on the internet based on the user's query. Information retrieval systems, both locally and globally, have been profoundly affected by the advent of the internet, and this includes the value of information left as comments on a page of SNS (Social Network Services). The concept of emotional and social connections is translated into a logical framework using the terms "node" and "link" to describe the structure of a social network. Efficient semantic models are more extensive training and evaluation materials, are needed to improve social network search capabilities. When compared with machine learning algorithms, the efficacy of traditional keyword-based search engines at understanding users' intentions is low. Recently, neural networks have gained popularity in the field of information retrieval because of their impressive vector representation learning capabilities. The usage of deep learning techniques for this purpose has recently been seen, and they have proven to be more effective than traditional machine learning techniques such artificial neural networks (ANNs). Improved results have been seen specifically using deep-learning techniques like long short-term memory (LSTM) & Recurrent Neural Network (RNN). In order to create a more personalized Information Retrieval(IR) system, this research suggests using a deep learning Hybrid RNN - LSTM model. Finally, the suggested method takes user comments into account and uses a hybrid RNN - LSTM to re-rank the data so that everyone is happy. Web search contest dataset is used for the implementation. Statistics like accuracy, precision, & recall are used to evaluate the data set on Bing and Duckduck go, two of the most prominent search engines. According to the findings, the proposed Hybrid method outperformed more traditional methods.