Schweizerischer Nationalfonds / Swiss National Fund
Artificial intelligence facilitates the detection of burnout. Scientists supported by the Swiss National Science Foundation have just developed a promising method based on automatic text analysis.
Burnout is a state of deep physical and mental fatigue. It is not easy to detect because its symptoms resemble those of illnesses such as depression or anxiety. But artificial intelligence opens up avenues to better recognize it: in an article recently published in Frontiers in Big Data
a team of scientists supported by the Swiss National Science Foundation (SNSF) presents a new method based on automatic language processing.
Currently, burnout is diagnosed through psychological tests that take the form of graded responses. For example: “I feel exhausted at the end of my working day: never/sometimes/every day”. However, such tests have significant limitations. For example, some people do not dare to tick the answers “never” and “every day” or are tempted to lie to influence the results.
More complete questionnaires, composed of open questions, can also be used to detect burnout. They deliver more relevant information but require significant analytical work. In practice, therefore, they are not applied.
A method based on text analysis
This is precisely what Mascha Kurpicz-Briki, professor of data engineering at the Bern University of Applied Sciences in Biel, wanted to remedy with her team. To do this, she used artificial intelligence in a method that automatically analyzes texts and identifies, on this basis, whether the language is burn-out or not. With success: the method correctly identifies 93% of cases of burnout. The scientist adds: “Automatic language processing is effective in detecting burnout while being less time-consuming, which is very promising.”
As part of this work, the scientist and her team analyzed texts from the Reddit platform – an English-speaking community website that functions as a discussion forum organized by theme. It has built up a database of more than 13,000 text extracts. Some came from discussions related to burnout while others came from various thematic forums.
Models trained with different data
She then used machine learning to develop a method that assesses whether a text is burnout or not. Concretely, she first classified the text extracts collected. The texts of threads about burnout were manually classified, in order to exclude those where burnout referred to something else. Texts from other threads, not related to mental health, have been labeled as not related to burnout. Based on these examples, she trained several models. Each used different configurations to determine whether a text – never seen by the model – contained burnout indications or not. These models were then pooled as part of the diagnostic method, which proved to be very effective.
Promising results that still need to be consolidated. The collaboration of medical experts is particularly necessary in a next step to verify the conclusions of this first exploratory project on real cases of burnout and on a representative sample of the population. The data collected on Reddit is indeed anonymous.
G. Merhbene, S. Nath, A. Puttick, M. Kurpicz-Briki: Burn-outEnsemble: Augmented Intelligence to Detect Indications for Burn-out in Clinical Psychology. Frontiers in Big Data (2022).
Rapid financing of original ideas
This project was supported by the Spark instrument of the SNSF. Spark aims to fund the testing or rapid development of new approaches, methods, theories, standards or ideas for scientific applications. Designed for projects with an unconventional concept and an original approach, it favors promising, audacious ideas based on very limited, if any, preliminary data. Spark is a pilot program launched by the SNSF for 2019-2020. It is currently in the evaluation phase to decide its future. ——————- The text of this press release and further information are available on the
from the Swiss National Science Foundation.
Contact:Mascha Kurpicz-Briki, Bern University of Applied Sciences, Höheweg 80, 2502 Biel, Tel.: +41 32 321 63 13, E-mail: email@example.com
We want to say thanks to the author of this short article for this outstanding web content
Tell me how you write, I’ll tell you if you’re burned out
Check out our social media accounts and also other related pageshttps://www.ai-magazine.com/related-pages/