Comparative Analysis of Pitching Prediction Algorithms

Two algorithms are described in the paper; one of them is the Kalman filter, which is based on the use of a pitching mathematical model, and the second uses a neural network in which the model is considered unknown. The results of the algorithms sensitivity analysis to the parameters of the model and its influence on the potential accuracy of prediction are presented. A stationary narrow-band second-order Markov process is used as a model of the ship pitching, which was used to form the input signal of the algorithms. Also, the results of the algorithms simulation in predicting real data are presented.

Improving the computation efficiency of polygenic risk score modeling: Faster in Julia

To enable large-scale application of polygenic risk scores in a computationally efficient manner we translate a widely used polygenic risk score construction method, Polygenic Risk Score – Continuous Shrinkage (PRS-CS), to the Julia programing language, PRS.jl. On nine different traits with varying genetic architectures, we demonstrate that PRS.jl maintains accuracy of prediction while decreasing the average run time by 5.5x. Additional programmatic modifications improve usability and robustness. This freely available software substantially improves work flow and democratizes utilization of polygenic risk scores by lowering the computational burden of the PRS-CS method.

Prediction of CPU Utilization in Cloud Environment during Seasonal Trend

Today, the most recent paradigm to emerge is that of Cloud computing, which promises reliable services delivered to the end-user through next-generation data centres which are built on virtualized compute and storage technologies Consumer will be able to access desired service from a “Cloud” anytime anywhere in the world on the bases of demand. Computing services need to be highly reliable, scalable, easy accessible and autonomic to support ever-present access, dynamic discovery and computability, consumers indicate the required service level through Quality of Service (QoS) parameters, according to Service Level Agreements (SLAs) A suitable mdel for the prediction is being developed. Here Genetic Algorithm is chosen in combination with stastical model to do the workload prediction .It is expected to give better result by producing less error rate and more accuracy of prediction compared to the previous algorithm.

Crop-coefficients of tomato as derived using monolithic weighing type lysimeter in mid hill region of Meghalaya

Crop coefficients (kc) was determined for tomato (Lycopersicon esculentum Mill.) with the help of UMS-GmBH cylindrical field lysimeter of 30 cm diameter and 120 cm deep and Penman-Monteith FAO-56 model. Eight other models viz. Modified Penman Method, Hargreaves equation, Samani-Hargreaves equation, Thornthwaite equation, Solar Radiation Method, Net Radiation Method, Blaney-Criddle Method and Radiation Method were also used for estimation of ET0­ and compared with Penman-Monteith model to find out the accuracy of prediction with limited weather parameters. Scatter plot and paired t-test were used for comparison. Out of all these models, Blaney-Criddle method, Solar and Net Radiation method were found to yield similar results as given by Penman-Monteith model. The values of crop evapo-transpiration (ETc) were varying from 2.54 mm d-1 to 6.70 mm d-1. The crop-coefficients (kc) for three growth stages of tomato viz., initial, mid and maturity were found to be 0.55, 1.07 and 0.78, respectively.

Ensemble learning for the early prediction of neonatal jaundice with genetic features

Background Neonatal jaundice may cause severe neurological damage if poorly evaluated and diagnosed when high bilirubin occurs. The study explored how to effectively integrate high-dimensional genetic features into predicting neonatal jaundice. Methods This study recruited 984 neonates from the Suzhou Municipal Central Hospital in China, and applied an ensemble learning approach to enhance the prediction of high-dimensional genetic features and clinical risk factors (CRF) for physiological neonatal jaundice of full-term newborns within 1-week after birth. Further, sigmoid recalibration was applied for validating the reliability of our methods. Results The maximum accuracy of prediction reached 79.5% Area Under Curve (AUC) by CRF and could be marginally improved by 3.5% by including genetic variant (GV). Feature importance illustrated that 36 GVs contributed 55.5% in predicting neonatal jaundice in terms of gain from splits. Further analysis revealed that the main contribution of GV was to reduce the false-positive rate, i.e., to increase the specificity in the prediction. Conclusions Our study shed light on the theoretical and practical value of GV in the prediction of neonatal jaundice.

RnkHEU: A Hybrid Feature Selection Method for Predicting Students’ Performance

Predicting students’ performance is one of the most concerned issues in education data mining (EDM), which has received more and more attentions. Feature selection is the key step to build prediction model of students’ performance, which can improve the accuracy of prediction and help to identify factors that have significant impact on students’ performance. In this paper, a hybrid feature selection method named rank and heuristic (RnkHEU) was proposed. This novel feature selection method generates the set of candidate features by scoring and ranking firstly and then uses heuristic method to generate the final results. The experimental results show that the four major evaluation criteria have similar performance in predicting students’ performance, and the heuristic search strategy can significantly improve the accuracy of prediction compared with forward search method. Because the proposed RnkHEU integrates ranking-based forward and heuristic search, it can further improve the accuracy of predicting students’ performance with commonly used classifiers about 10% and improve the precision of predicting students’ academic failure by up to 45%.

Nanotechnology-Based Sensitive Biosensors for COVID-19 Prediction Using Fuzzy Logic Control

Increasing the growth of big data, particularly in healthcare-Internet of Things (IoT) and biomedical classes, tends to help patients by identifying the disease early through methods for the analysis of medical data. Hence, nanotechnology-based IOT biosensors play a significant role in the medical field. Problem. However, the consistency continues to decrease where missing data occurs in such medical data from nanotechnology-based IOT biosensors. Furthermore, each region has its own special features, which further lowers the accuracy of prediction. The proposed model initially reconstructs lost or partial data in order to address the challenge of handling the medical data structures with incomplete data. Methods. An adaptive architecture is proposed to enhance the computing capabilities to predict the disease automatically. The medical databases are managed by unpredictable environments. This optimized paradigm for diagnosis produces the fuzzy, genetically categorized decision tree algorithm. This work uses a normalized classifier namely fuzzy-based decision tree (FDT) algorithm for classifying the data collected via nanotechnology-based IOT biosensors, and this helps in the identification of nondeterministic instances from unstructured datasets relating to the medical diagnosis. The FDT algorithm is further enhanced by using genetic algorithms for effective classification of instances. Finally, the proposed system uses two larger datasets to verify the predictive precision. In order to describe a fuzzy decision tree algorithm based upon the fitness function value, a modified decision classification rule is used. The structure and unstructured databases are configured for processing. Results and Conclusions. This evaluation of test patterns helps to track the efficiency of FDT with optimized rules during the training and testing stages. The proposed method is validated against nanotechnology-based IOT biosensors data in terms of accuracy, sensitivity, specificity, and F -measure. The results of the simulation show that the proposed method achieves a higher rate of accuracy than the other methods. Other metrics relating to the model with and without feature selection show an improved sensitivity, specificity, and F -measure rate than the existing methods.

Short term wind power interval prediction based on VMD and improved BLS

Aiming at the problem of wind power interval prediction, a short-term wind power interval prediction model based on VMD and improved BLS is proposed. Firstly, the complex wind power time series are decomposed by variational mode decomposition to reduce the non stationarity of wind power. Then an improved broad learning system (BLS) is established to predict the power and error of each component, and a weight is given to the prediction error of each component. The sparrow search algorithm (SSA) is used to optimize the weight, and the width of the prediction interval is obtained by adding the power and error prediction values. The experimental data show that the proposed model improves the accuracy of prediction interval.