(Press-News.org) A growing number of academic researchers are mining social media data to learn about both online and offline human behavior. In recent years, studies have claimed the ability to predict everything from summer blockbusters to fluctuations in the stock market.
But mounting evidence of flaws in many of these studies points to a need for researchers to be wary of serious pitfalls that arise when working with huge social media data sets, according to computer scientists at McGill University in Montreal and Carnegie Mellon University in Pittsburgh.
Such erroneous results can have huge implications: thousands of research papers each year are now based on data gleaned from social media. "Many of these papers are used to inform and justify decisions and investments among the public and in industry and government," says Derek Ruths, an assistant professor in McGill's School of Computer Science.
In an article published in the Nov. 28 issue of the journal Science, Ruths and Jürgen Pfeffer of Carnegie Mellon's Institute for Software Research highlight several issues involved in using social media data sets - along with strategies to address them. Among the challenges:
Different social media platforms attract different users - Pinterest, for example, is dominated by females aged 25-34 - yet researchers rarely correct for the distorted picture these populations can produce.
Publicly available data feeds used in social media research don't always provide an accurate representation of the platform's overall data - and researchers are generally in the dark about when and how social media providers filter their data streams.
The design of social media platforms can dictate how users behave and, therefore, what behavior can be measured. For instance, on Facebook the absence of a "dislike" button makes negative responses to content harder to detect than positive "likes".
Large numbers of spammers and bots, which masquerade as normal users on social media, get mistakenly incorporated into many measurements and predictions of human behavior.
Researchers often report results for groups of easy-to-classify users, topics, and events, making new methods seem more accurate than they actually are. For instance, efforts to infer political orientation of Twitter users achieve barely 65% accuracy for typical users - even though studies (focusing on politically active users) have claimed 90% accuracy.
Many of these problems have well-known solutions from other fields such as epidemiology, statistics, and machine learning, Ruths and Pfeffer write. "The common thread in all these issues is the need for researchers to be more acutely aware of what they're actually analyzing when working with social media data," Ruths says.
Social scientists have honed their techniques and standards to deal with this sort of challenge before. "The infamous 'Dewey Defeats Truman' headline of 1948 stemmed from telephone surveys that under-sampled Truman supporters in the general population," Ruths notes. "Rather than permanently discrediting the practice of polling, that glaring error led to today's more sophisticated techniques, higher standards, and more accurate polls. Now, we're poised at a similar technological inflection point. By tackling the issues we face, we'll be able to realize the tremendous potential for good promised by social media-based research."
INFORMATION:
"Social Media for Large Studies of Behavior," Ruths and Pfeffer, Science, Nov. 28, 2014.
American presidents spend their time in office trying to carve out a prominent place in the nation's collective memory, but most are destined to be forgotten within 50-to-100 years of their serving as president, suggests a study on presidential name recall released today by the journal Science.
"By the year 2060, Americans will probably remember as much about the 39th and 40th presidents, Jimmy Carter and Ronald Reagan, as they now remember about our 13th president, Millard Fillmore," predicts study co-author Henry L. Roediger III, PhD, a human memory expert at Washington ...
High-tech genomics and traditional Chinese medicine come together as researchers identify the genes responsible for the intense bitter taste of wild cucumbers. Taming this bitterness made cucumber, pumpkin and their relatives into popular foods, but the same compounds also have potential to treat cancer and diabetes.
"You don't eat wild cucumber, unless you want to use it as a purgative," said William Lucas, professor of plant biology at the University of California, Davis and coauthor on the paper to be published Nov. 28 in the journal Science.
That bitter flavor in ...
Human-induced changes to Earth's carbon cycle - for example, rising atmospheric carbon dioxide and ocean acidification - have been observed for decades. However, a study published this week in Science showed human activities, in particular industrial and agricultural processes, have also had significant impacts on the upper ocean nitrogen cycle.
The rate of deposition of reactive nitrogen (i.e., nitrogen oxides from fossil fuel burning and ammonia compounds from fertilizer use) from the atmosphere to the open ocean has more than doubled globally over the last 100 years. ...
MEDFORD/SOMERVILLE, Mass.(November 27, 2014, 2 PM) -- New catalysts designed and investigated by Tufts University School of Engineering researchers and collaborators from other university and national laboratories have the potential to greatly reduce processing costs in future fuels, such as hydrogen. The catalysts are composed of a unique structure of single gold atoms bound by oxygen to several sodium or potassium atoms and supported on non-reactive silica materials. They demonstrate comparable activity and stability with catalysts comprising precious metal nanoparticles ...
People affected by a common inherited form of autism could be helped by a drug that is being tested as a treatment for cancer.
Researchers who have identified a chemical pathway that goes awry in the brains of Fragile X patients say the drug could reverse their behavioural symptoms.
The scientists have found that a known naturally occurring chemical called cercosporamide can block the pathway and improve sociability in mice with the condition.
The team at the University of Edinburgh and McGill University in Canada identified a key molecule - eIF4E - that drives excess ...
Led by a University of Oklahoma professor, an international team of researchers has discovered the first evidence of milk consumption in the ancient dental calculus--a mineralized dental plaque--of humans in Europe and western Asia. The team found direct evidence of milk consumption preserved in human dental plaque from the Bronze Age to the present day.
"The study has far-reaching implications for understanding the relationship between human diet and evolution," said Christina Warinner, professor in the OU Department of Anthropology. "Dairy products are a very recent, ...
Researchers have discovered a mechanism linked to the brain damage often suffered by stroke victims--and are now searching for drugs to block it.
Strokes happen when the blood supply to part of the brain is cut off but much of the harm to survivors' memory and other cognitive function is often actually caused by "oxidative stress" in the hours and days after the blood supply resumes.
A team from the University of Leeds and Zhejiang University in China studied this second phase of damage in laboratory mice and found a mechanism in neurons that, if removed, reduced the ...
A large-scale study led by the University of Exeter has found that boys are more likely to be stillborn than girls. Published in the journal BMC Medicine, the study reviewed more than 30 million births globally, and found that the risk of stillbirth is about ten percent higher in boys. This equates to a loss of around 100 000 additional male babies per year.
The results could help to explain why some pregnancies go wrong. Around a quarter of stillbirths have no known cause. Of the remainder, many are linked to placental abnormalities but it is often unclear why the abnormalities ...
Drinking 3-5 cups of coffee per day may help to protect against Alzheimer's Disease, according to research highlighted in an Alzheimer Europe session report published by the Institute for Scientific Information on Coffee (ISIC), a not-for-profit organisation devoted to the study and disclosure of science related to coffee and health.
The number of people in Europe aged over 65 is predicted to rise from 15.4% of the population to 22.4% by 20251 and, with an aging population, neurodegenerative diseases such as Alzheimer's Disease are of increasing concern. Alzheimer's Disease ...
Using world-leading research methods, the team of Dr David Wright and Prof Paul Holmes, working with Dr Jacqueline Williams from the Victoria University in Melbourne, studied activity in an area of the brain responsible for controlling movements when healthy participants observed a video showing simple hand movements and simultaneously imagined that they were performing the observed movement.
Using transcranial magnetic stimulation - a technique where a coil placed over the scalp delivers a stimulation to the brain, activates neurons in the underlying area, and causes ...
