Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/336252 
Autor:innen: 
Erscheinungsjahr: 
2025
Quellenangabe: 
[Journal:] Pakistan Journal of Commerce and Social Sciences (PJCSS) [ISSN:] 2309-8619 [Volume:] 19 [Issue:] 4 [Year:] 2025 [Pages:] 673-713
Verlag: 
Johar Education Society, Pakistan (JESPK), Lahore
Zusammenfassung: 
Current studies primarily focus on the recommender systems (RSs) through the algorithmic accuracy, and which insufficient to fully measure accurately the practical effectiveness of the RSs and understand user experience. There is an urge examine the dynamic interactions on the critical user context of "beyond accuracy" features such as diversity, novelty, and serendipity to understand RS's value creation mechanism holistically. The aim of the study is to explore the multidimensional effects of key features (Stimuli) of artificial intelligence (AI)-RSs on consumer behavior (Response) based on the Stimulus-Organism-Response (S-O-R) theoretical framework which is a gap in the literature. The study was modeled four key features of RSs (accuracy, diversity, novelty, serendipity) as stimuli, user engagement, satisfaction, and perceived risk as organism (mediating constructs), and behavioral intentions as the response. Data was collected from 437 participants, and analyses were conducted using the covariance-based structural equation modeling (CB-SEM) by AMOS 24.0 and SPSS 25.0 statistical programs. The findings explains that the empirical assumptions of the S-O-R model were met to a high degree, successfully explaining 61% of the variance in behavioral intentions. Specifically, satisfaction and engagement have strongest positive effects on behavioral intentions. The findings suggest that the effectiveness of AI-RSs should no longer be evaluated solely through algorithmic accuracy, but rather through the dynamics of user-centered value creation (satisfaction, engagement, and risk management). Furthermore, serendipity have the strongest direct effect on engagement, highlighting the importance of "beyond-accuracy objectives" that trigger curiosity. Accuracy and diversity significantly and negatively reduce risk, demonstrating a critical role in increasing system trustworthiness. Contrary to the accuracy-diversity trade-off problem, a strong positive relationship exists between accuracy and diversity, suggesting that these multiple goals reinforce each other in the context of user perception. These results provide important theoretical and practical consequences for the design of multi-objective recommender systems (MORS).
Schlagwörter: 
Accuracy
AI-Powered Recommendation Systems
Diversity
Engagement
Netflix viewers
Novelty
Perceived Risk
Satisfaction
Serendipity
Türkiye
Persistent Identifier der Erstveröffentlichung: 
Creative-Commons-Lizenz: 
cc-by-nc Logo
Dokumentart: 
Article

Datei(en):
Datei
Größe
1.04 MB





Publikationen in EconStor sind urheberrechtlich geschützt.