20.500.12556/DKUM-95861 Učinkovitost klepetalnih robotov, ki temeljijo na umetni inteligenci na področju mentalnega zdravja The effectiveness of artificial intelligence-based chatbots in the field of mental health Uvod: Klepetalni roboti, ki temeljijo na umetni inteligenci, s svojo obdelavo naravnega jezika predstavljajo potencialno koristno orodje za podporo osebam s težavami na področju mentalnega zdravja. Cilj tega pregleda je bil oceniti učinkovitost uporabe klepetalnih robotov na področju mentalnega zdravja. Metode: Iskanje je potekalo v štirih mednarodnih podatkovnih bazah (CINAHL Ultimate, Medline, PubMed in Web of Science) z ustreznim iskalnim nizom. Pridobljeni članki so bili nato ustrezno analizirani in sintetizirani. Kritična presoja zbranih dokazov je potekala po hierarhiji dokazov. Kritična ocena člankov je izvedena s pomočjo ocenjevalnega orodja JBI. Rezultati: Od 492 zadetkov je 13 člankov izpolnjevalo vse pogoje, od katerih smo v končno analizo vključili tri sistematične preglede RCT in 10 študij RCT. Ugotovljeno je bilo, da so se intervencije s klepetalnimi roboti izkazale za učinkovite pri izboljšanju mentalnega zdravja. Učinkovitost klepetalnih robotov je bila v največji meri zaznana pri zmanjševanju simptomov depresije, anksioznosti in stresa, posledično z izboljšanjem simptomov drugih motenj mentalnega zdravja. Razprava in zaključek: Klepetalni roboti so se izkazali za učinkovite pri zmanjševanju simptomov motenj mentalnega zdravja. Za dodatno potrditev teh rezultatov bo treba opraviti še več raziskav o uporabi klepetalnih robotov na področju mentalnega zdravja in njihovem prihodnjem vključevanju v prakso. Introduction: AI-based chatbots, using natural language processing, represent potentially valuable tools for supporting individuals with mental health problems. The aim of this review was to evaluate the effectiveness of chatbot interventions in the field of mental health. Methods: A search was conducted in four international databases (CINAHL Ultimate, Medline, PubMed, and Web of Science) using a relevant search strategy. The retreived articles were then analyzed and synthesized accordingly. A critical apraisal of the collected evidence was carried out according to the hierarchy of evidence. The JBI critical appraisal tool was used to assess the quality of the studies. Results: The search yielded 492 articles, of which only 13 met all inclusion criteria. These included three systematic reviews of RCTs and 10 individual RCTs that were incorporated into the final analysis. The findings indicated that chatbots interventions were effective in improving mental health. The effectiveness was most evident in reducing symptoms of depression, anxiety, and stress, which also contributed to alleviating symptoms of other mental health disorders. Discussion and conclusion: Chatbots have proven effective in reducing symptoms of mental health disorders. However, further research is needed to confirm these results and to explore the future integration of chatbots into mental health practice. klepetalni roboti mentalno zdravje umetna inteligenca chatbots mental health artificial intelligence true false false [M. Gajić] Slovenski jezik Angleški jezik Magistrsko delo/naloga 2025-11-03 10:59:54 2025-12-16 11:54:36 2025-12-17 03:14:48 0000-00-00 00:00:00 2025 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 613.86:007.52+004.8(043.2) 259939331 MAG_Gajic_Marijana_2025.pdf MAG_Gajic_Marijana_2025.pdf 1 2A525BF4C83BABBF84303AD336752FA5 56fb87234e8d42ce10c82860bc989eed99c0b61fabd5f9a1b6721336dfdabc7d 794c18cb-b89d-11f0-b13b-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=200329 Fakulteta za zdravstvene vede 0 0 0