Abstract

Childhood trauma is a recognized precursor to mental disorders (Kerns et al., 2015). This study investigates the extensive impact of childhood and adolescent trauma on social, interpersonal, and psychological health, referencing the works of Keane & Barlow (2004), Koss et al. (1991), and Kulka et al. (1990). Utilizing domestic violence reports from Taiwan spanning 2006 to 2022, we analyzed correlations with child violence rates during the corresponding years. Employing machine learning for predictive analysis, our findings indicate a troubling trend: victims of domestic violence frequently become perpetrators or experience social disabilities.
Previous research establishes that adverse childhood experiences—including sexual abuse, physical and emotional neglect, familial mental illness, and instability due to parental divorce or separation—are strongly linked to negative developmental outcomes (Neofytou, 2022). Maxfield (2004) found that chronic childhood trauma increases the likelihood of violent behavior by over 200%, further exacerbating developmental challenges and perpetuating cycles of violence.
Our comprehensive analysis of open-source domestic violence data from Taiwan reveals a significant positive correlation with child violence rates. Using the ARIMA model (5, 1, 0) for time series prediction, we forecast an 8.59% increase in domestic violence in Taiwan over the next three years, with a mean absolute percentage error (MAPE) of 8.25%, maintaining overall prediction accuracy within 10%. Notably, during the COVID-19 pandemic (2019-2021), cases of child violence surged by 20% and remain elevated.
Furthermore, we leverage electrocardiogram (ECG) and heart rate analysis to detect emotional changes in victims, facilitating early intervention and prevention of further victimization. This interdisciplinary approach integrates engineering and medical technologies, advocating for a proactive social and medical focus on prevention and early treatment to reduce recovery time and minimize hospital visits.