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

Numerical evaluation of cytologic data. IV. Discrimination and classification

P H Bartels

    Analytical and Quantitative Cytology
    |March 1, 1980
    PubMed
    Summary

    This study introduces methods for assigning data to categories using classification rules. Linear discriminant functions and probability density-based approaches are explored to reduce classification errors and define decision boundaries.

    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

    Global acetylation and methylation changes predict papillary urothelial neoplasia of low malignant potential recurrence: a quantitative analysis.

    International journal of immunopathology and pharmacology·2011
    Same author

    A PC-based system for the objective analysis of histologic specimens through quantitative contextual karyometry.

    Applied optics·2010
    Same author

    Knowledge engineering in quantitative histopathology.

    Applied optics·2010
    Same author

    Self-optimizing, self-learning system in pictorial pattern recognition.

    Applied optics·2010
    Same author

    Early stromal invasion in cervical cancer: immunocytochemical changes occurring in infiltrating neoplastic cells.

    Advances in clinical pathology : the official journal of Adriatic Society of Pathology·2007
    Same author

    Karyometry in the early detection and chemoprevention of intraepithelial lesions.

    European journal of cancer (Oxford, England : 1990)·2005

    Area of Science:

    • Machine Learning
    • Statistical Classification
    • Data Analysis

    Background:

    • Categorical data assignment necessitates robust classification rules.
    • Existing methods may have limitations in accuracy and decision boundary complexity.
    • Optimizing classification accuracy is crucial for data interpretation.

    Purpose of the Study:

    • To explore and demonstrate methods for data classification.
    • To compare linear discriminant functions with probability density-based classification.
    • To illustrate techniques for reducing classification errors.

    Main Methods:

    • Utilizing linear discriminant functions for classification.
    • Applying variance weighting to discriminant functions to minimize errors.
    • Employing probability density estimation for nonlinear decision boundaries.
    • Demonstrating approaches with numerical examples for bivariate data.

    Main Results:

    • Linear discriminant functions offer computationally efficient classification rules.
    • Weighting discriminant functions by data set variances can decrease classification errors.
    • Probability density-based classification results in nonlinear decision boundaries.
    • Numerical examples validate the effectiveness of both approaches.

    Conclusions:

    • Linear discriminant functions provide a practical approach to data categorization.
    • Variance weighting enhances the performance of linear discriminant methods.
    • Probability density methods offer a more flexible, albeit complex, alternative for classification.
    • The choice of method depends on data characteristics and desired decision boundary complexity.

    Related Experiment Videos