Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Estimating the posterior probabilities using the k-nearest neighbor rule.

Amir F Atiya1

  • 1Department of Computer Engineering, Cairo University, Giza, Egypt. amir@alumni.caltech.edu

Neural Computation
|April 2, 2005
PubMed
Summary

This study introduces a novel posterior probability estimator for pattern classification. The method uses weighted K-nearest neighbors, improving classification confidence and accuracy.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Novel pricing strategies for revenue maximization and demand learning using an exploration-exploitation framework.

Soft computing·2021
Same author

Epileptic Seizures Detection Using Deep Learning Techniques: A Review.

International journal of environmental research and public health·2021
Same author

Handling of uncertainty in medical data using machine learning and probability theory techniques: a review of 30 years (1991-2020).

Annals of operations research·2021
Same author

A Novel Active Learning Regression Framework for Balancing the Exploration-Exploitation Trade-Off.

Entropy (Basel, Switzerland)·2020
Same author

A Bias and Variance Analysis for Multistep-Ahead Time Series Forecasting.

IEEE transactions on neural networks and learning systems·2015
Same author

Comprehensive review of neural network-based prediction intervals and new advances.

IEEE transactions on neural networks·2011

Area of Science:

  • Machine Learning
  • Pattern Recognition
  • Statistical Modeling

Background:

  • Accurate posterior probability estimation is crucial for classification tasks requiring confidence measures.
  • Existing methods may not sufficiently capture nuanced class probabilities.

Purpose of the Study:

  • To propose a novel posterior probability estimator for pattern classification.
  • To enhance confidence measures in classification outcomes.

Main Methods:

  • The proposed estimator utilizes K-nearest neighbors (KNN).
  • Weights are assigned to each neighbor, contributing additively to the probability estimate.
  • Neighbor weights, summing to 1, are estimated via maximum likelihood.

Main Results:

  • Simulation studies demonstrate the effectiveness of the proposed estimator.
  • The method provides improved posterior probability estimates compared to standard approaches.

Conclusions:

  • The novel weighted KNN posterior probability estimator offers a valuable tool for pattern classification.
  • This approach enhances the reliability of classification confidence assessment.

Related Experiment Videos