A novel approach to text document embedding inhigh dimensional spaces for information Retrieval inLarge Databases

for many applications such as document classification, information retrieval, and machine translation. he representations accuracy of text document information will ave broad applications in healthcare analytics. In the existing ystem, the text document information and query have been represented without conducting the query understanding. In this ork, we propose text document embedding in high dimensional pace using the Meta information, which learns distributed epresentations of health care datasets. Relational networks are sed to represent the domain-specific Meta information to handle earch challenges. The developed relation network use the predefined ext document information processed based on vector pace, term weights calculated using TF-IDF method. The roposed method conduct the information understanding by inding the relationship between the text documents and achieve he search accuracy.