A paper published in the journal Science suggests that handwriting analysis can be more accurate than you might expect
New Scientist article Written by Simon LeBlancA study of handwriting analysis, published online in the peer-reviewed journal Science, suggests that it can be a more accurate tool than previously thought.
The research, led by the University of Cambridge’s John Lacey, used a type of analysis called handwriting recognition to track the evolution of letters in handwriting, which is a notoriously difficult task.
The results showed that using a technique called “penumbra” (writing) analysis to determine a person’s handwriting was more accurate and accurate than previously believed.
“Our analysis showed that handwriting recognition accuracy was comparable to that of face-recognition,” said Lacey.
“In other words, we were able to accurately recognise faces and their signatures.
This is a significant result in the area of handwriting recognition because it indicates that the process of recognising handwriting is not as hard as previously thought.”
The study was funded by the Science Foundation of India and the British Science Foundation.
It’s the first study to look at handwriting analysis as a more general tool for the development of handwriting detection, the researchers said.
Previous research has shown that handwriting has a number of important properties, including that it is more accurate when handwriting is written using the right hand, and more accurate if it is written with the left hand.
These properties make handwriting recognition very difficult, and have been used to develop a number different types of handwriting-based detection systems.
The study’s findings also suggest that handwriting can be used to detect other handwriting, and that a handwriting analysis system can also be used as a way of identifying and classifying people, or even to predict the future.
“A lot of the research is based around the concept that handwriting is a highly specific kind of writing and that you need to have the right kind of tool to do this,” said LeBlanch.
“This work, by contrast, showed that we can also use handwriting to tell us things about people’s personality, such as their history and personality traits.”
The paper is the latest in a series of studies that have explored handwriting in the context of other types of information.
In a study published last year, researchers at Stanford University found that handwriting and facial recognition have similar properties in that they are “correlated”.
“Facial recognition and handwriting are both relatively similar, so there’s a strong correlation between both,” said study co-author Liza Gittleson, the postdoctoral researcher who led the Stanford study.
“We found that the best predictor of handwriting is whether you’re looking at facial recognition.
The best predictor for handwriting is the handwriting.
There’s also a correlation between handwriting and personality.”
Gittles on the difference between handwriting, facial recognition and personality:The researchers compared the handwriting and the personality traits of 1,829 participants who had completed a questionnaire, which asked them to provide some details about their personality and their handwriting.
“The handwriting was the least predictive, but it also had the least correlations with personality,” Gittgeson said.
“They’re not the same thing.
People who have good handwriting tend to be kinder and kinder, and those who have bad handwriting tend not to be friendly and kind.”
To test the relationship between handwriting as a tool for recognising people and their personalities, the Stanford researchers looked at the handwriting of more than 4,000 people from the UK and the US.
The researchers found that “correlation” was a very strong predictor of the handwriting in those participants who were more likely to be found with the right handwriting, or who were not overly influenced by their handwriting-related preferences.
“Correlation is a very powerful predictor of personality, and handwriting is one of the few things that we know can predict personality,” said Gitteson.
The authors also looked at handwriting from individuals who had received training on the task, and found that participants who performed the best in their handwriting recognition tended to have a more positive attitude towards their handwriting than did those who were the least successful.
This was true for both men and women, and also for people who had an education and were self-taught.
“People who had a more confident personality tended to be more likely than people who didn’t have that confidence to write well,” said the study’s lead author, Simon Leblanc.
“It’s interesting that handwriting recognisers are also quite good at predicting personality, but the personality of handwriting recognizers is not the only predictor.
Personality is also correlated with handwriting recognition.”
LeBlanch and Gittmeson are currently working on a paper about handwriting recognition that will be published later this year.
The University of Warwick has a research and technology centre that provides research funding for the University.
For more information about handwriting, visit the University’s website: www.wsu.ac.uk/writing/faculty-and-staff/fac_fac_corp_gittlesons_gumby.aspx.
To find out more about the University, visit its website