We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
A Modified Approach for Data Retrieval for Identifying Primary Causes of Deaths.
- Authors
Noman, A. H. M.; Das, Kumer; Andrei, Stefan
- Abstract
Automatic data retrieval ensures fast, secure and accurate data extraction from structured and unstructured documents. This study introduces a modified approach of retrieving data from the World Health Organization (WHO) database. WHO maintains a large mortality database based on age, sex, and cause of death of various countries all over the world. An Integrated Development Environment (IDE) of programming language R, called RStudio, has been used to develop this retrieval process. There are over 2,000 front-line, user-contributed packages available on the Comprehensive R Archive Network (CRAN), a network of File Transfer Protocol (FTP) and web servers around the world. It stores identical, up-to-date, versions of code and documentation for R. The RCurl package offers high-level facilities in R to communicate with Hypertext Transfer Protocol (HTTP) servers. It is very useful to download Uniform Resource Locators (URLs) and submit forms in various ways. R has built-in data frame to store data tables. The dplyr package is used to work with the data frames. After data retrieval, the Principal Component Analysis (PCA) and Singular Value Decomposition (SVD) have been applied to find out the major causes of deaths in WHO mortality database. The objective of this research is twofold. Firstly, to retrieve data automatically from large database websites in such a way that it does not rely on any prior knowledge about the target pages and their contents, and secondly, to apply data dimension reduction techniques to identify primary causes of deaths.
- Subjects
INFORMATION retrieval; UNIFORM Resource Locators; CAUSES of death; HTTP (Computer network protocol); SINGULAR value decomposition; INTERNET servers
- Publication
ACET Journal of Computer Education & Research, 2020, Vol 14, Issue 1, p1
- ISSN
1547-3716
- Publication type
Article