A Parallelized Influenced Citation Analysis for Medical Documents using Modified BioBERT or ClinicalBERT Model Semantic Similarity Measure
Main Article Content
Abstract
Doctors have a limited amount of time, which is quite valuable. They can be wasting their time by reading a variety of publications that have the same semantic substance. Since it emphasizes the importance of the study, gives credit to the researchers, and helps prevent semantic repetition, citation analysis is a crucial part of medical research,and assists in avoiding semantic repetition. Citations are often used by researchers for a variety of reasons, but the primary purpose has to do with demonstrating how the references have impacted their work. These days, the majority of references have a tendency to have some kind of impact on research. The suggested technique takes a semantic approach rather than depending just on keyword matching in order to discover impacted citations in medical research articles. This is accomplished by eliminating the need for keyword matching. References or a bibliography are often included at the conclusion of each and every publication that pertains to medical study. The purpose of this model is to determine, within a semantic framework, the degree of relevance that exists between the medical research document and the list of reference papers that it contains or among given research papers.
In this study, the evolution of the suggested work is described in two different ways: the first is the introduction of a new semantic similarity measure that is used to compute impacted citation scores. This measure is calculated by using Modified BioBERT or ClinicalBERT. Up to this point, every single semantic metric has been calculated by means of using concurrent execution. We are doing this by leveraging parallel algorithms in order to make the process of detecting semantic similarity more efficient. In the end, the work that is being offered is able to efficiently identify affected citations in medical research publications by using high-performance computing (HPC).