The AI system used Natural Language Processing (NLP) to analyze speech patterns in a sample of older people.

Researchers at the University of California San Diego School of Medicine have found in a study that AI tools using NLP can predict the level of loneliness in older adults.

In fact, rapid advances in AI, machine learning, and NLP have raised the possibility of detection of mental conditions such as psychosis, PTSD, bipolar disorder, and depression by an analysis of the patient’s natural speech. (NEW ATLAS)

Loneliness

It’s a condition that is particularly difficult to quantify in the absence of any objective measures. Moreover, verbal interviews often result in bias because the subject is reluctant to admit the condition.

Ellen Lee, the senior author of the new research, said: “For this project, we used natural language processing or NLP, an unbiased quantitative assessment of expressed emotion and sentiment, in concert with the usual loneliness measurement tools.”

For the study, the researchers evaluated 80 older adults using long, conversational, and semi-structured interviews that stretched to as long as 90 minutes. They also applied conventional assessments used for the condition.

Observations

• The lonelier a person, the longer their responses to questions on the subject of loneliness
• Likely presence of a typical “lonely speech” pattern
• Men used more fearful and joyful words
• Women were more likely to clearly express feelings of loneliness

The system could qualitatively predict a person’s loneliness with 94% accuracy.

Potential

Though the subject study is only proof-of-concept, in the future AI could help (in real-time) people to identify the loneliness within them, and to take positive actions to remedy it.

Related Story:       AI-Powered Apps Can Now Ease a Divorce