prediction

Diabetes is a chronic disease associated with abnormalhigh levels of glucose in the blood. Diabetesmakemany kinds of complications, which also leads to a high rate of repeated admission of patients with diabetes. The goal of this study is to Predict hospital readmission of Diabetic patients with machine learningtechniques.

The data used in the study are data obtained from the UCI machine learning repository about diabetic patients. The dataset used contains 100,000 instances and it include 55 features from 130 hospitals in the United States for10 years.

This articlegets results from the final stages of evaluation. In this evaluation process, compared the performance of decision tree, random forest, Xgboost, k-neighbors, Adaboost and deep neural network withaccuracy.

Thenumber of selected features by PCA-based feature selection method improve the predictive performance based on accuracy of deep learning and most machine learning models for predicting readmission. The improvement of machine learning models depended on the specific choice of the prediction model, number of selected features, and “k” for k-fold validation.

An evaluation of the relationship between crop yield and soil properties would be useful in estimating the fluctuations in yield, and in an implementation of correct field management. The study was conducted in farmer operated wheat fields in Sorkhankalateh district, 25 km northeast of Gorgan, Golestan province, Iran. Soil samples (0-30 cm depth) were collected just after crop planting at the end of autumn 2004 from a 100 × 180m plot as a nested grid (n=101). A 1 m2 plot of wheat was harvested at each 101 previously sampled sites at the end of spring. Statistical results showed that frequency distribution of the data was normal. ESP had the variability of CV=12.36% while pH was of the lowest variability (CV=0.59%). For principal component analysis, 7 principal components were used in the study. Selection criterion then was employed for explaining the effective parameters in each component. The eigenvector for each PC, therefore was selected on the basis of having a value larger than the SC value. The results showed that such soil fertility parameters as available P (0.847), OM% (0.810), total N (0.742), available K (0.727) and CEC (0.725) bore larger loadings and therefore had the major role in soil variability. The results suggest that soil and crop yield variability were affected by management. The results of principle component analysis indicated that variability within the field was mostly derived from fertility parameters with multi-regression models explaining 57% of total variability of the yield in wheat.

This study has been conducted to investigate the predictability of stock returns behavior of the companies listed on "Tehran Stock Exchange" and also to predict the stock returns by using "Artificial Neural Networks". In order to predict the returns, in the first stage, the historical data relating to time series of the companies plus three variables of technical analysis (stock index, volume of exchanged stock and the closed price) for a five year period (July 1998 - 2003) were used. The optimum model for prediction of stock daily return for each company was designed via changes in parameters of the artificial neural network. In the second stage, prediction of daily return during the same period was done based only on past information or historical process of time series. In this study, "Multi Layered Perceptron(MLP)"artificial neural network with different learning functions were used.