Dissertation > Excellent graduate degree dissertation topics show

The Balanced m×2Cross-validation Method

Author: DuWeiJie
Tutor: LiJiHong
School: Shanxi University
Course: Probability Theory and Mathematical Statistics
Keywords: Cross-validation Balanced m×2cross-validation Model selection
CLC: O212.1
Type: Master's thesis
Year: 2013
Downloads: 29
Quote: 0
Read: Download Dissertation

Abstract


Estimating expected prediction error of statistical model is one of the core tasks of statistical machine learning. The performance of expected prediction error estimation has a significant impact on model selection, test of significance of different model prediction errors and variable selection. To find a good estimation, researchers proposed a large number of estimating methods, such as MDL, Cross-validation (CV), Bootstrap and.632Bootstrap. Cross-validation is a widely used generalization error estimating method.This paper presents an analysis and summary of the advantages and disadvantages of existed CV methods on machine learning classification tasks, and improves the CV method on this basis. Although the Blocked3x2CV has the characteristics of low fold number, balanced dataset segmentation and small number of experiments,5x2CV and10x2CV also have good effect on classification. Furthermore, for the2-fold CV, the experimental results will be improved to some extent with the increase of frequency. But how to solve the contradiction between the performance of experiments and experimental expense is always a difficult issue in the field of machine learning. We promote and improve the Blocked3x2CV, advanced Balanced7x2CV and Balanced11x2CV. Based on these results, we sum up more generic Balanced mx2CV, at the same time propose the construction method of Balanced mx2CV. We determine the experiments times with the Balanced mx2CV, which improves the experiment performance, then support our opinion with theoretical analysis and simulation experiments.We use the Balanced mx2CV method on classification model selection tasks. Considering various influence factors of the Balanced mx2CV model selection methods, we can see that the Balanced mx2CV model selection methods is better than5-fold CV and10-fold CV. At the same time, it has been proved that the Balanced mx2CV has the same consistency of selection as the standard CV. Thus, it states that taking the Balanced m×2CV on the classification tasks is more appropriate than Random m×2CV.

Related Dissertations

  1. Bayesian methods under two ordinal value of multi-value data model with simultaneous identification of outliers,O212.1
  2. Model Selection for Longitudinal Data Based on the Quadratic Inference Function under a Simple Order Restriction,O212.1
  3. The Cumulative logistic Regression Classification of Students’ Poverty Data,O212.1
  4. Design and Development of Slewing Hydraulic System on 100t Crawler Crane,TH213.7
  5. Research on the Techno-Economic Analysis and Evaluation Method in Waste Heat Power Station,TM617
  6. Random Network Model Discrimination,O157.5
  7. Automatic Target Recognition fluorescent magnetic particle inspection image processing techniques,TP391.41
  8. Research on Human Meridian Potential Signal Processing and Physical Status Classification Method,TN911.7
  9. Currently, the Group centralized capital management, risk control Issues,F275
  10. Study on Application of Support Vector Regression in Combined Forecast,F224
  11. Research on a Number of Issues Related to Partial Least Squares Regression Analysis,O212.1
  12. Hunan social security tax reform,F812.42;F224
  13. The Study of Computer-aided Diagnosis System of Eye Disease Based on Pattern Recognition Technology,TP399-C8
  14. Research on Blind Equalization Algorithm Based on Support Vector Regressor,TN911.5
  15. The Bayes Factor Estimation of the Order for Stationary AR Model,O212.8
  16. Nonlinear modeling of neural-network construction and selection,TP183
  17. Applying Study on Time Series Modeling and Model Selection,F224
  18. A Practical Implementation of the Gibbs Sampler for Mixture of Distributions,O212.2
  19. Positioning the interaction of genes statistical methods,Q3-33
  20. The Further Researches on Some EOQ Models,F224
  21. Selection and Realization of Objective Model of Administrative Procedural Law,D915.4

CLC: > Mathematical sciences and chemical > Mathematics > Probability Theory and Mathematical Statistics > Mathematical Statistics > General mathematical statistics
© 2012 www.DissertationTopic.Net  Mobile