Researchers have developed a data-sorting algorithm which claims to rank science literature by usefulness. A team at North Carolina State University created a text mining algorithm for a genomics database which ranks papers selected for inclusion by relevancy and novelty. The algorithm can only be applied to toxicogenomics, a specialized field which studies how toxic chemicals interact with genes[/url], but similar algorithms can be created for other fields.
The algorithm is intended as a work aide for scientists. “Over 33,000 scientific papers have been published on heavy metal toxicity alone, going as far back as 1926,” said North Carolina State's Allan Peter Davis. “We simply can’t read and code them all. And, with the help of this new algorithm, we don’t have to.”