This well documented commentary goes hand in hand with the very recently published Science article (much shorter by necessity).
We believe that such news items contain important information from the viewpoint of investors, but they are much shorter than complete news articles.
As described in Sect. 1, patient records are much shorter than full-text biomedical articles, such that patient record is a weak indicator of relevance, thus our feedback-based CDS method is expected to outperform the method (BM25+Sim_{d_{Para}text Q}).
A monograph is expected to cover all the aspects of the study concerned in detail (average word count approx. 50 000 words); in contrast, journal articles are much shorter (approximately 3000 words), less detailed and cover only certain aspects of the study.
As the patient records are usually much shorter than the full-text biomedical articles, they do not necessarily contain sufficient amount of semantic evidence of relevance.
> Since this task addresses the classification of sentences which are usually much shorter than paragraphs or the whole articles, the bag-of-words features from a local sentence tend to have high risk of data sparseness (18), which could result in a biased representation for low-frequency words and degrade the classification performance.
A possible explanation is that the documents in traditional newswire or Web collections are in general much shorter and more coherent than the scientific articles.
Most people may not have noticed that the front-page nameplate is higher, or that the articles inside are more numerous and much shorter.
Second, most existing works dealwith news articles or medical reports which are generally much shorter than Supreme Courts judgments.
Magazine and newspaper articles cover a variety of subjects, and they're generally much shorter than reading an entire book.
