Data Mining-Based Stability and Prescription Analysis of Neonatal Parenteral Nutrition Solution

In order to evaluate the stability of neonatal parenteral nutrition solution, in this paper, the prescription of neonatal parenteral nutrition solution was investigated and analyzed. The formula of neonatal parenteral nutrition solution used, particularly the one utilized in this study, is commonly used in clinical practice. All the neonatal parenteral nutrition solution required for the test was prepared on the purification workbench in a sterile environment. The time points of stability of parenteral nutrient solution were 0, 12, and 24 hours, respectively, and three parallel samples were taken at each time point. Likewise, to investigate the stability of two kinds of fat milk injection in parenteral nutrition solution of neonates and provide a reference for subsequent experiments and to investigate the influence of electrolyte, amino acid, temperature, pH value, mixing sequence, and the final concentration of glucose on the stability of neonatal parenteral nutrition solution, the stability indexes of neonatal parenteral nutrition liquid mainly include appearance, pH, insoluble particles, fat milk particle size, and particle size distribution. Neonatal parenteral nutrition solution prescriptions from the First Affiliated Hospital of Jinan University, specifically from January to June 2019, were collected and statistically processed. The experimental data were processed by SPSS 19.0 software and data mining technology. The results were expressed as mean ± standard deviation and statistically processed by ANOVA. P < 0.05 was considered statistically significant. The results showed that the stability of neonatal parenteral nutrient solution was influenced by many factors. The formula of neonatal parenteral nutrition solution is generally reasonable, but there are unreasonable phenomena which are needed to be improved further if feasible.

Online regularized pairwise learning with non-i.i.d. observations

In this paper, we consider the online regularized pairwise learning (ORPL) algorithm with least squares loss function for non-independently and identically distribution (non-i.i.d.) observations. We first establish new Bennett’s inequalities for [Formula: see text]-mixing sequence, geometrically [Formula: see text]-mixing sequence, [Formula: see text]-geometrically ergodic Markov chain and uniformly ergodic Markov chain. Then we establish the convergence rates for the last iterate of the ORPL algorithm with the polynomially decaying step sizes and varying regularization parameters for non-i.i.d. observations. These established results in this paper extend the previously known results of ORPL from i.i.d. observations to the case of non-i.i.d. observations, and the established result of ORPL for [Formula: see text]-mixing can be nearly optimal rate of ORPL for i.i.d. observations with [Formula: see text]-norm.