Oscillatory perturbations with varying amplitudes and frequencies have been found to significantly affect human standing balance. However, previous studies have only applied perturbation in either the anterior-posterior (AP) or the medio-lateral (ML) directions. Little is currently known about the impacts of 2D oscillatory perturbations on postural stability, which are more commonly seen in daily life (i.e., while traveling on trains, ships, etc.). This study investigated the effects of applying 2D perturbations vs 1D perturbations on standing stability, and how increasing the frequency and amplitude of perturbation impacts postural stability. A dual-axis robotic platform was utilized to simulate various oscillatory perturbations and evaluate standing postural stability. Fifteen young healthy subjects were recruited to perform quiet stance on the platform. Impacts of perturbation direction (i.e., 1D versus 2D), amplitude, and frequency on postural stability were investigated by analyzing different stability measures, specifically AP/ML/2D Center-of-Pressure (COP) path length, AP/ML/2D Time-to-Boundary (TtB), and sway area. Standing postural stability was compromised more by 2D perturbations than 1D perturbations, evidenced by a significant increase in COP path length and sway area and decrease in TtB. Further, the stability decreased as 2D perturbation amplitude and frequency increased. A significant increase in COP path length and decrease in TtB were consistently observed as the 2D perturbation amplitude and frequency increased. However, sway area showed a considerable increase only with increasing perturbation amplitude but not with increasing frequency.

We attempted to apply a novel approach to stock market predictions. The Logistic Regression machine learning algorithm (Joseph Berkson) was applied to analyze news article headlines as represented by a bag-of-words (tri-gram and single-gram) representation in an attempt to predict the trends of stock prices based on the Dow Jones Industrial Average. The results showed that a tri-gram bag led to a 49% trend accuracy, a 1% increase when compared to the single-gram representation’s accuracy of 48%.