Similarity Identification of Large-scale Biomedical Documents using Cosine Similarity and Parallel Computing

Document similarity computation is an important research topic in information retrieval, and it a crucial issue for automatic document categorization. The value between 0 1, then the closest to 1 represented both documents considered more relevant, vice versa. However, large scale of textual has created problem finding relevance level documents. Therefore, mesh heading text PubMed higher than abstract Furthermore, parallel computing implemented speed up large-scale identification process that automatically calculates application. execution time 15.447 seconds, timely 74.191 seconds. because contains words heading. This study successfully identified biomedical cosine algorithm. result shown texts form graph table useful measure based on TF*IDF calculation result.