Parental Parenting Competence Assessment using Multilayer Backpropagation Neural Network Model

Parental parenting ability assessment is at the core of child development and family functioning. In this research, a multilayer Backpropagation (BP) Neural Network model is suggested for parental parenting competence assessment, overcoming the shortcomings of conventional methods like qualitative questionnaires and expert scores, which are time-consuming and subjective. Conventional machine learning algorithms such as decision trees and support vector machines fail to identify the intricate, non-linear patterns in parenting behavior. The proposed BP Neural Network model processes multidimensional parenting data—such as emotional support, discipline styles, communication skills, and parent involvement—providing a holistic assessment and data-driven approach. In contrast to other traditional machine learning methods, the BP model exhibits stronger deep feature extraction, increasing evaluation accuracy and minimizing bias. Experimental results based on common parenting datasets indicate the model reaches an accuracy of 92.4% and an F1-score of 91.8%, which is higher than traditional approaches. The research advances the domain of intelligent psychological assessment through the development of a more accurate, scalable, and objective methodology for assessing parenting skills, assisting professionals such as psychologists, teachers, and policymakers in designing effective parenting interventions.