Toward automatic evaluation of medical abstracts: The current value of sentiment analysis and machine learning for classification of the importance of PubMed abstracts of randomized trials for stroke

The exponential growth of the number of medical publications makes it more and more difficult for the clinical practitioner to keep track of relevant scientific progress. Computer-aided screening and analysis might help to overview and classify the content of medical publications more efficiently. The new wave of enthusiasm for machine learning algorithms has launched several new possibilities to analyze and categorize documents, ranging from sports news to hospital discharge summaries. Search algorithms for the medical literature, i.e.
Source: Journal of Stroke and Cerebrovascular Diseases - Category: Neurology Authors: Source Type: research