This computer program read 700,000 journal entries to understand what makes humans angry, happy or sad

Artificial intelligence can tell us more about the mental health benefits of looking after ourselves.

Researchers from Canada’s University of Waterloo gathered more than 700,000 anonymous online journal entries written by over 67,000 users of a mobile mood tracking app. They then developed an AI computer model that could identify keywords within the text.

What makes us feel happy?

The report, published in January, demonstrated a strong correlation between positive moods and sleep quality, healthy eating and exercising. In short, the more we look after ourselves, the happier we seem to be.

The research also highlighted correlations between moods themselves. For example, those who reported feeling tired were more likely to experience sadness, anger and frustration. While those who reported feeling productive were more likely to experience feelings of happiness and ecstasy.

Even “productive” activities such as getting a haircut were linked to feeling calmer and happier.

Link between mental and physical health

We’ve known for some time that mental and physical health are closely linked, but there is still plenty of work to be done to help us understand the correlation.

Research like this highlights the link between looking after ourselves – for example, by eating healthily, sleeping better and exercising – and our psychological well-being.

Lukasz Golab, who supervised the study, praised the technology at the heart of the research, saying it could eventually be used as a type of “screening tool” to flag possible mental health issues among social media users.

According to the World Health Organization (WHO), nearly one in five of the world’s children and adolescents have a mental disorder. Depression is the most prevalent, affecting a total of 254 million people around the world.

Happiness and artificial intelligence

This isn’t AI’s first journey to try and uncover the secret to happiness. A study in 2019 used an AI model to analyze 8 million books and 65 million newspaper articles, creating an “index of national happiness” that went back to 1820.

Among the key findings were that longer lives mean happier lives, increased national income leads to increased national happiness and that war can, unsurprisingly, dramatically lower the national mood.