The latest research report shows that artificial intelligence (AI) can detect loneliness from a person’s speech with an accuracy rate of 94%. American researchers use IBM Watson Multiple AI tools analyze the loneliness of the elderly interviewed.
By analyzing the words, phrases, and silence gaps in interviews, artificial intelligence’s assessment of loneliness symptoms in the elderly is almost as accurate as the results they fill in the report questionnaire. At the same time, artificial intelligence also shows that lonely people usually answer direct questions about loneliness. Longer, express more sadness in the answer .
According to Ellen Lee, senior author of the research report and the University of California, San Diego School of Medicine, most studies either directly ask: “How long will it take you to feel lonely?” This may lead to biased reactions related to loneliness. In this research project, we use natural language processing, which is an unbiased quantitative assessment of expressed emotions and emotions, combined with the usual loneliness measurement tools.
The interesting thing about this tool is that it not only uses dictionary-based methods, such as searching for specific words that reflect fear, but also presents corresponding patterns by testing the words used in the response.
Experts pointed out that there has been a “loneliness epidemic” in the United States in recent years, which is characterized by rising suicide rates and opioid use rates, declining productivity, increasing medical costs, and increasing mortality. A California report published earlier this year Research by the University of San Diego showed that 85% of residents living in independent elderly communities experience moderate to severe loneliness.
The COVID-19 pandemic and the ensuing lockdown of the city have increased people’s time alone and made the situation worse. Researchers want to know more about how natural language processing technology and machine learning models predict the loneliness of elderly people living in the community.
This study focused on 80 independent living residents between the ages of 66 and 94. Their average age was 83. Between April 2018 and August 2019 (before the outbreak of the new crown epidemic), trained researchers were The tester conducted a semi-structured interview.
The tester was asked 20 questions. These questions came from the UCLA Loneliness Scale, which uses a four-level rating scale to answer some questions, such as: Do you often feel ignored by others? Do you often feel that you are part of a group of friends?
Subjects were also interviewed in private conversations, which were recorded and manually transcribed. Then use natural language processing tools including IBM Watson Natural Language Understanding Software (WNLU) to test the transcription to quantify and express emotions.
The first author of the research report and University of California, San Diego Valsa Badar said that the WNLU software system uses deep learning to extract metadata from keywords, categories, emotions, and grammar. Natural language models and machine learning enable us to systematically Examine long-term interviews from multiple testers and explore how subtle language features such as emotions express feelings of loneliness.
He also pointed out that similar human sentiment analysis may be divergent, lack of consistency, and require extensive training to be standardized. Compared with the scores of UCLA’s Loneliness Scale, using language features, the artificial intelligence system can predict loneliness with an accuracy of 94%.
The accuracy rate of artificial intelligence predicting self-recognition of loneliness was 94%, while the accuracy rate of “quantified loneliness (based on the scores of the UCLA Lonely Scale)” was 76%. They found that lonely people were in personal interviews It takes longer to answer questions and expresses more sadness when answering direct questions about loneliness.
The study also pointed out that there are differences between men and women. Women are more likely than men to admit that they feel lonely in the test. Compared with women, men use more words of fear and joy in their responses, which indicates that they are negative The experience with positive emotions is more extreme, and even shows that men can express these emotions more freely.
Ellen pointed out that when older women and men directly answer questions describing loneliness, there are subtle gender differences in their emotions and emotional expression. This study emphasizes the difference between the research assessment of loneliness and the subjective experience of loneliness by testers, and artificial intelligence systems can help identify this.
Researchers claim that there may be “loneliness language” that can be used to detect the loneliness of the elderly, which will improve the true evaluation of the elderly by clinicians and family members, thereby helping to treat their loneliness, especially in the epidemic During the outbreak closure.
Currently, the University of California, San Diego, is exploring the natural language pattern characteristics of loneliness and wisdom. These characteristics are negatively correlated among the elderly, which means that the higher the intelligence of the elderly, the stronger the loneliness. The co-author of the research report, Dilip Jester of UCSD, said: “Language data can be combined with our other assessments of cognition, movement, sleep, physical activity, and mental health to improve our understanding of aging. Cognition and help us live a healthy old age.”
The study compared the accuracy of artificial intelligence with the tester’s own loneliness report. As the research pointed out, loneliness does not always reflect real feelings and emotions. However, artificial intelligence and self-reports can be psychologically affected. The combination of scientists and professionals can improve the accuracy of diagnosis.
Ellen said: “We agree that the UCLA Loneliness Scale has some inaccuracies because it relies on self-reports. However, the Loneliness Scale is one of the most popular tools because it does not explicitly use’loneliness’. This word, and it seems that it can always capture the characteristics of loneliness without gender bias. We hope to develop more accurate tools to assess people’s loneliness.” The latest research report is published in the recent publication of “America’s Elderly” Journal of Psychiatry.
For more such interesting article like this, app/softwares, games, Gadget Reviews, comparisons, troubleshooting guides, listicles, and tips & tricks related to Windows, Android, iOS, and macOS, follow us on Google News, Facebook, Instagram, Twitter, YouTube, and Pinterest.