Can a Siri-like chatbot help lonely seniors?

Nearly One-Quarter of Americans Never Plan on Retiring

It's not quite the Joaquin Phoenix-Scarlett Johansson love story from 2013 romance "Her," but researchers believe an artificial intelligence chatbot may help relieve human loneliness, particularly in seniors.

» RELATED: Why are Americans so lonely? Massive study finds nearly half of US feels alone, young adults most of all

Computing scientist Osmar Zaïane of Canada's Alberta Machine Intelligence Institute envisions "a device that's emotionally intelligent, where an elderly person can say 'I'm tired' or 'It's beautiful outside,' or tell a story about their day—and receive a response that carries on the conversation and keeps them engaged," he said in a statement.

Zaïane and his colleagues created an AI model able to express responses with emotions, though some emotions such as surprise and love were easier to express compared to others.

Once the technology is able to understand and match an expressed emotion (by converting speech to text), the next step is to teach the program to “independently decide on what emotion to express, depending on the persona it’s talking to,” Zaïane said.

» RELATED: Loneliness increases your risk of Type 2 diabetes, study suggests

Loneliness among Canada’s elderly has been called a public health crisis. According to Statistics Canada, as many as 1.4 million elderly Canadians report being lonely, a condition that commonly leads to depression and overall deterioration in health.

Loneliness is also a major issue in America, where more than one-third of older adults report feeling lonely, the AARP Foundation noted in its 2018 survey. The report suggests maintaining meaningful relationships can significantly help.

“Companionship—a cat, a dog, other people—helps tremendously,” Zaïane echoed. “The advantage for caregivers of a digital companion like this is it can also collect information on the emotional state of the person, noting if they are frequently feeling sad, for example.”

The study was published in April in the 5th International Conference on Internet Science.