(Press-News.org)
Word lists are the basis of so much research in so many fields. Researchers at the Complexity Science Hub have now developed an algorithm that can be applied to different languages and can expand word lists significantly better than others.
Many projects start with the creation of a word list. Not only in companies when mind maps are created, but also in all areas of research. Imagine you want to find out on which days people are in a particularly good mood by analyzing Twitter postings. Just looking for the word "happy" wouldn't be enough.
Instead, you would have to use an algorithm that detects all tweets that indicate that someone is happy. "So the first step is to create a list of all the words that indicate just that. The whole research stands or falls on doing so," explains Anna Di Natale, a researcher at the Complexity Science Hub in Vienna. But how to come up with the most accurate, complete word lists possible?
A PROBLEM THAT CONCERNS MANY
This widespread problem not only concerns opinion researchers who want to find out how politicians' statements are received by the public. Companies, too want to find out how their products are perceived through sentiment analysis.
To improve things, Di Natale has now developed a new method, called LEXpander, that outperforms previous algorithms. And this even in two different languages - German and English. Moreover, for the very first time ever, she has developed a way through which it is possible to compare different tools at all.
IMPROVED PERFORMANCE
In comparison with four other algorithms for wordlist expansion (WordNet, Empath 2.0, FastText and GloVe), LEXpander performed significantly better, especially in German. For example, the researchers found that LEXpander guesses 43% of words right when expanding an English word list for positive meaning. A very popular model, FastText, in comparison, is right only 28% of the time.
INDEPENDENCE FROM THE LANGUAGE ITSELF
The reason is that this tool works language-independently. It is not based on one language, but on a so-called colexification network. This recognized linguistic concept resides on homonyms and polysemies, single words that have two or more distinct meanings. For example: the ancient Greek word φάρμακον (pharmacon) can mean medicine or poison. Two different things, but thematically close. But there are others that don't suggest kinship - such as "bank" as a financial institution or the land alongside a river.
"If you collect them across many languages - and here we analyzed about 19 different languages - you can see connections between them," Di Natale says. The network is formed when these colexifications occur in several languages across different language families, creating connections.
This independence from the language itself allows LEXpander to achieve better results in different languages. "There are many methods developed for English. They work very well and quickly and everyone uses them. Trying to apply them to other languages works, but not as well as it might work if you had started developing a method for German or Italian,” Di Natale explains.
IMPORTANT FOR NEW TOPICS LIKE COVID
For many topics there are already good word lists. But for new topics - like when COVID came up - new ones have to be created. Until now, they were usually created by hand during brainstorming with colleagues and several tools were used to help. But until now there was no way to compare them. Anna Di Natale and her team have now created this possibility and have also developed a new tool that performs better than the others. This can be an important cornerstone for many future research projects in various fields.
FIND OUT MORE
The study “LEXpander: Applying colexification networks to automated lexicon expansion” has been published in Behavior Research Methods.
ABOUT THE COMPLEXITY SCIENCE HUB
The mission of the Complexity Science Hub (CSH Vienna) is to host, educate, and inspire complex systems scientists dedicated to making sense of Big Data to boost science and society. Scientists at the Complexity Science Hub develop methods for the scientific, quantitative, and predictive understanding of complex systems.
The CSH Vienna is a joint initiative of AIT Austrian Institute of Technology, Central European University CEU, Danube University Krems, Graz University of Technology, IIASA, Medical University of Vienna, TU Wien, VetMedUni Vienna, Vienna University of Economics and Business, and Austrian Economic Chambers (WKO). https://www.csh.ac.at
END
Yersinia bacteria cause a variety of human and animal diseases, the most notorious being the plague, caused by Yersinia pestis. A relative, Yersinia pseudotuberculosis, causes gastrointestinal illness and is less deadly but naturally infects both mice and humans, making it a useful model for studying its interactions with the immune system.
These two pathogens, as well as a third close cousin, Y. enterocolitica, which affects swine and can cause food-borne illness if people consume infected meat, have many traits in common, particularly their knack for interfering with the immune system’s ability to respond to infection.
The plague pathogen is blood-borne and transmitted by infected ...
Hydrogen’s potential as a clean fuel could be limited by a chemical reaction in the lower atmosphere, according to research from Princeton University and the National Oceanic and Atmospheric Association.
This is because hydrogen gas easily reacts in the atmosphere with the same molecule primarily responsible for breaking down methane, a potent greenhouse gas. If hydrogen emissions exceed a certain threshold, that shared reaction will likely lead to methane accumulating in the atmosphere — with decades-long climate consequences.
“Hydrogen is theoretically the fuel of the future,” said Matteo Bertagni, a postdoctoral researcher at the High ...
BOSTON – A type of immune therapy called chimeric antigen receptor (CAR)-T cell therapy has revolutionized the treatment of multiple types of blood cancers but has shown limited efficacy against glioblastoma—the deadliest type of primary brain cancer—and other solid tumors.
New research led by investigators at Massachusetts General Hospital (MGH) and published in the Journal for ImmunoTherapy of Cancer March 10, 2023, suggests that drugs that correct abnormalities in a solid tumor’s blood vessels can improve the delivery and function of CAR-T cell therapy.
With CAR-T cell therapy, immune cells are taken from a ...
DETROIT – A team of faculty from Wayne State University has discovered new technology that will quickly and easily detect active Mycobacterium tuberculosis (TB) infection antibodies. Their work, “Discovery of Novel Transketolase Epitopes and the Development of IgG-Based Tuberculosis Serodiagnostics,” was published in a recent edition of Microbiology Spectrum, a journal published by the American Society for Microbiology. The team is led by Lobelia Samavati, M.D., professor in the Center for Molecular Medicine and Genetics in the School of Medicine. ...
AMHERST, Mass. – University of Massachusetts Amherst biologist Lynn Adler has won the Mahoney Life Sciences Prize for her research demonstrating that different kinds of wildflowers can have markedly different effects on the health and reproduction rate of bumblebees.
“My lab studies the role that flowers play in pollinator health and disease transmission,” says Adler. “Flowers are of course a food source for pollinators, but, in some cases, nectar or pollen from specific plants can be medicinal. However, flowers are also high-traffic areas, and just like with humans, high-traffic areas can be hotspots for disease transmission. We’re tracing how different populations ...
New research from Washington University in St. Louis provides evidence that modern gender norms and biases in Europe have deep historical roots dating back to the Middle Ages and beyond, suggesting that DNA is not the only thing we inherit from our ancestors.
The findings — published on March 13, 2023 in the Proceedings of the National Academy of Sciences (PNAS) — highlight why gender norms have remained stubbornly persistent in many parts of the world despite significant strides made by the international ...
Populations of whooper swans grow 30 times faster inside nature reserves, new research shows.
Whooper swans commonly spend their winters in the UK and summers in Iceland.
In the new study, researchers examined 30 years of data on swans at 22 UK sites – three of which are nature reserves managed by the Wildfowl and Wetlands Trust (WWT).
Survival rates were significantly higher at nature reserves, and population growth was so strong that many swans moved to non-protected sites.
Based on these findings, the research team – led by the universities of Exeter and Helsinki – project that nature reserves could help double the number ...
The cycling of carbon through the environment is an essential part of life on the planet.
Understanding the various sources and reservoirs of carbon is a major focus of Earth science research. Plants and animals use the element for cellular growth. It can be stored in rocks and minerals or in the ocean. Carbon in the form of carbon dioxide can move into the atmosphere, where it contributes to a warming planet.
A new study led by Florida State University researchers found that plants and small organisms in Arctic rivers could be responsible for more than half the particulate organic matter flowing to the Arctic Ocean. That’s a significantly ...
NEW YORK, NY (March 13, 2023)--A new study by Columbia University researchers suggests that cholesterol-lowering drugs called statins have the potential to reduce heart disease in people with obstructive sleep apnea regardless of the use of CPAP machines during the night.
CPAP (continuous positive airway pressure) therapy improves sleep quality and reduces daytime fatigue in people with obstructive sleep apnea. But based on findings from several recent clinical trials, CPAP does not improve heart health as physicians originally hoped.
Alternative ...
New research suggests that around 0.5% of global carbon emissions could be captured during the normal crushing process of rocks commonly used in construction, by crushing them in CO2 gas.
The paper ‘Mechanochemical processing of silicate rocks to trap CO2’ published in Nature Sustainability says that almost no additional energy would be required to trap the CO2. 0.5% of global emissions would be the equivalent to planting a forest of mature trees the size of Germany.
The materials and construction industry ...
