Analysis of Factors Affecting Seasonal Affective Disorder in Public Environment and College Football Players Based on Computerized Neural Network Light Therapy

Abstract

Seasonal affective disorder is a depressive affective psychiatric disorder that recurs at the same time of the year and seriously impacts people's daily work and life. The university stage is an important period for individuals to transition to social life, in which they are more vulnerable to negative life events such as academic performance pressure, interpersonal discomfort, and employment problems. Hence, the incidence of depression among university football players is at a high level. As an important timing factor, ambient light has a wide range of effects on various physiological and psychological functions, and its non-visual effects on mood have attracted particular attention from researchers. The illuminance, color temperature and wavelength of ambient light are important physical factors influencing mood. Abnormal light patterns such as short photoperiods, artificial light at night, and continuous light can lead to mood disorders. Light duration, time point, individual characteristics, subjective preferences, and genotype also modulate the mood effects of light. On the one hand, light signals are projected by intrinsic light-sensitive ganglion cells in the retina to brain regions involved in emotion regulation to directly influence mood. On the other hand, light signals indirectly influence mood by synchronizing internal physiological rhythms and their regulated hormone secretion, neurotransmission and sleep. The proposed method uses heart rate, exercise behavior, environment, and textual information from social platforms as raw data for mental health analysis; feature extraction of various types of information by convolutional neural network in artificial intelligence; and random forest algorithm as a classifier to determine the factors influencing seasonal affective disorder in college football players. The test and data analysis results show that the scheme described in the paper has a high recognition accuracy, which proves the effectiveness and feasibility of the scheme.