Empowering Digital Parenting: Evaluating Performance and Usability of Mobile Recommendation Systems for Parent Education
Technology, Knowledge and Learning, 2026 (ESCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1007/s10758-026-10014-7
- Dergi Adı: Technology, Knowledge and Learning
- Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus, IBZ Online, Aerospace Database, Agricultural & Environmental Science Database, Applied Science & Technology Source, Compendex, EBSCO Education Source, Education Abstracts, Educational research abstracts (ERA), ERIC (Education Resources Information Center), INSPEC, Psycinfo, Social Science Premium Collection (ProQuest), Education Collection (ProQuest), Education Source Ultimate (EBSCO), Technology Collection (ProQuest)
- Anahtar Kelimeler: Child, Digital parenting skills, Machine learning, Mobile recommendation system, Parent, Performance, Usability criteria
- Gazi Üniversitesi Adresli: Evet
Özet
Recommender systems are systems that make recommendations to people based on their behavior, preferences, wants and needs. Since the recommendations are based on real user behaviors, they are effective in enabling users to make direct decisions based on the recommendations and are always up to date. These systems, which support individualization of education, enable individuals to develop their own strategies in learning. Therefore, recommendation systems have become inevitable for the effective use of mobile devices that have become widespread in all areas of our lives. Especially considering the limited time of adults and their abandonment of online learning environments, these systems are more important for adult education. Considering all these, it was decided to use mobile recommendation systems to increase the digital parenting competencies of adults. However, it is necessary to ensure the continuity of the use of the developed digital parenting recommendation system by users and to prioritize user satisfaction. For this reason, this study aims to examine the performance (accuracy, sensitivity, precision and F1 score) and usability of the digital parenting recommendation system. In line with this purpose, answers to the following research questions are sought. In this study, developmental research Type-1 research model was used. Algorithm sensitivity calculations for the models produced in the mobile recommendation system were calculated for each of the output variables of opening items, playing videos, commenting, sharing content, rating content and adding to playlist. These calculations were performed for decision trees, k-nearest neighbor, support vector machines, naive bayes classifier and logistic regression analysis algorithms. The results showed that, on a scale ranging from 0 to 1, the decision trees achieved the highest sensitivity values for item unpacking (0.70), video playback (0.77), content rating (0.70), and playlist insertion (0.71). For the output variables of commenting (0.51) and sharing (0.82), the algorithm with the highest sensitivity was naive bayes classifiers.