News by Subject Social & Behavior

Researchers from the Vrije Universiteit Amsterdam and the Dutch Royal Academy's Humanities Cluster have performed a thorough evaluation of four different name recognition tools on popular 40 novels, including A Game of Thrones. Their analyses, published in PeerJ Computer Science, highlight types of names and texts that are particularly challenging for these tools to identify as well as solutions for mitigating this.

We are more original than we think -- this is what is being suggested by literary text analysis carried out by a new method of stylometry proposed by scientists from the Institute of Nuclear Physics Polish Academy of Sciences. The author's individuality can already be seen in connections between no more than a dozen of words in English text. It turns out that in Slavic languages authorship identification requires even fewer words and is more certain.

Patient matching is crucial for patient safety and care. First of it's kind research from the Regenstrief Institute, Indiana University and Pew Charitable Trusts has found that standardizing last names and address information can improve linking of patient health records by as much as 8 percent -- more than 2 billion records -- and enhance data available to clinicians to inform their care decisions.

ORNL used artificial intelligence to analyze data from published medical studies about bullying to reveal the potential of broader impacts, such as mental illness or disease; New low-cost, printed, flexible sensor can wrap around power cables to precisely monitor electrical loads from household appliances; ORNL is evaluating paths for licensing remotely operated microreactors, which could provide clean energy to isolated regions; ORNL-led team used carbon nanotubes to improve process that removes salt from water.