Large Minibatch Training on Supercomputers with Improved Accuracy and Reduced Time to Train

2018 ◽  
Author(s):  
Valeriu Codreanu ◽  
Damian Podareanu ◽  
Vikram Saletore
Keyword(s):  
Improved Accuracy ◽  
Reduced Time

Ultrasound guided thyroid FNA with cytopathology in-attendance significantly improved accuracy in a university teaching hospital

2015 ◽  
Author(s):  
Mohamed Ahmed ◽  
Michael Jeffers ◽  
John Feeney ◽  
Pardeep Govender ◽  
Mark Sherlock ◽  
...  
Keyword(s):  
Teaching Hospital ◽  
University Teaching ◽  
Ultrasound Guided ◽  
Thyroid Fna ◽  
Improved Accuracy

Unifying Calibration Parameters for Dielectric Moisture Meter

2006 ◽  
Vol 2 (1) ◽  
pp. 73-94 ◽  
Author(s):  
Péter Mészáros ◽  
David B. Funk
Keyword(s):  
The United States ◽  
Temperature Correction ◽  
Physical Parameters ◽  
Correction Coefficient ◽  
Prototype System ◽  
United States Department ◽  
Very High Frequency ◽  
Grain Moisture ◽  
University Of Budapest ◽  
Improved Accuracy

The Unified Grain Moisture Algorithm is capable of improved accuracy and allows the combination of many grain types into a single “unified calibration”. The purposes of this research were to establish processes for determining unifying parameters from the chemical and physical properties of grains. The data used in this research were obtained as part of the United States Department of Agriculture-Grain Inspection, Packers and Stockyards Administration's Annual Moisture Calibration Study. More than 5,000 grain samples were tested with a Hewlett-Packard 4291A Material/Impedance Analyzer. Temperature tests were done with a Very High Frequency prototype system at Corvinus University of Budapest. Typical chemical and physical parameters for each of the major grain types were obtained from the literature. Data were analyzed by multivariate chemometric methods. One of the most important unifying parameters (Slope) and the temperature correction coefficient were successfully modeled. The Offset and Translation unifying parameters were not modeled successfully, but these parameters can be estimated relatively easily through limited grain tests.

Computer Modeling of a Tire under Cornering Loads

1989 ◽  
Vol 17 (2) ◽  
pp. 86-99 ◽  
Author(s):  
I. Gardner ◽  
M. Theves
Keyword(s):  
Finite Element Analysis ◽  
Geometric Approach ◽  
Lateral Force ◽  
Slip Angle ◽  
Element Analysis ◽  
Pressure Distributions ◽  
Slip Effects ◽  
Improved Accuracy ◽  
Nonlinear Geometric ◽  
Good Agreement

Abstract During a cornering maneuver by a vehicle, high forces are exerted on the tire's footprint and in the contact zone between the tire and the rim. To optimize the design of these components, a method is presented whereby the forces at the tire-rim interface and between the tire and roadway may be predicted using finite element analysis. The cornering tire is modeled quasi-statically using a nonlinear geometric approach, with a lateral force and a slip angle applied to the spindle of the wheel to simulate the cornering loads. These values were obtained experimentally from a force and moment machine. This procedure avoids the need for a costly dynamic analysis. Good agreement was obtained with experimental results for self-aligning torque, giving confidence in the results obtained in the tire footprint and at the rim. The model allows prediction of the geometry and of the pressure distributions in the footprint, since friction and slip effects in this area were considered. The model lends itself to further refinement for improved accuracy and additional applications.

Algorithm Comparation of Naive Bayes and Support Vector Machine based on Particle Swarm Optimization in Sentiment Analysis of Freight Forwarding Services

2020 ◽  
Vol 4 (2) ◽  
pp. 362-369
Author(s):  
Sharazita Dyah Anggita ◽  
Ikmah
Keyword(s):  
Support Vector Machine ◽  
Sentiment Analysis ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Support Vector ◽  
The Public ◽  
Svm Algorithm ◽  
Bayes Algorithm ◽  
Freight Forwarding ◽  
Improved Accuracy

The needs of the community for freight forwarding are now starting to increase with the marketplace. User opinion about freight forwarding services is currently carried out by the public through many things one of them is social media Twitter. By sentiment analysis, the tendency of an opinion will be able to be seen whether it has a positive or negative tendency. The methods that can be applied to sentiment analysis are the Naive Bayes Algorithm and Support Vector Machine (SVM). This research will implement the two algorithms that are optimized using the PSO algorithms in sentiment analysis. Testing will be done by setting parameters on the PSO in each classifier algorithm. The results of the research that have been done can produce an increase in the accreditation of 15.11% on the optimization of the PSO-based Naive Bayes algorithm. Improved accuracy on the PSO-based SVM algorithm worth 1.74% in the sigmoid kernel.

Bioluminescent proteins prediction with voting strategy.

2020 ◽  
Vol 15 ◽  
Author(s):  
Shulin Zhao ◽  
Ying Ju ◽  
Xiucai Ye ◽  
Jun Zhang ◽  
Shuguang Han
Keyword(s):  
Amino Acid ◽  
Model Building ◽  
Prediction Method ◽  
Pair Composition ◽  
Proposed Model ◽  
Counting Rules ◽  
Significant Phenomenon ◽  
Voting Strategy ◽  
Improved Accuracy ◽  
Selection Of

Background: Bioluminescence is a unique and significant phenomenon in nature. Bioluminescence is important for the lifecycle of some organisms and is valuable in biomedical research, including for gene expression analysis and bioluminescence imaging technology.In recent years, researchers have identified a number of methods for predicting bioluminescent proteins (BLPs), which have increased in accuracy, but could be further improved. Method: In this paper, we propose a new bioluminescent proteins prediction method based on a voting algorithm. We used four methods of feature extraction based on the amino acid sequence. We extracted 314 dimensional features in total from amino acid composition, physicochemical properties and k-spacer amino acid pair composition. In order to obtain the highest MCC value to establish the optimal prediction model, then used a voting algorithm to build the model.To create the best performing model, we discuss the selection of base classifiers and vote counting rules. Results: Our proposed model achieved 93.4% accuracy, 93.4% sensitivity and 91.7% specificity in the test set, which was better than any other method. We also improved a previous prediction of bioluminescent proteins in three lineages using our model building method, resulting in greatly improved accuracy.

Sign in / Sign up

Export Citation Format

Share Document