Contact
Classification with Similarity Metrics
3071
portfolio_page-template-default,single,single-portfolio_page,postid-3071,bridge-core-2.4.3,ajax_fade,page_not_loaded,,qode-title-hidden,qode_grid_1300,side_area_uncovered_from_content,overlapping_content,qode-content-sidebar-responsive,qode-theme-ver-22.8,qode-theme-bridge,disabled_footer_top,wpb-js-composer js-comp-ver-6.3.0,vc_responsive,elementor-default,elementor-kit-3194

Topic

Classification

Classifiers, which are studied in AI and cognitive science, often employ similarity metrics

Setup

SNF

Part of the Swiss National Science Foundation (SNF) Doc.CH. Project New Approaches to Cognition

Design

R

Realized in R-Statistics, Python, and C++

Classification with Similarity Metrics

Finding out which similarity metric leads to human-like classification performance.

Classifying objects—assigning them to a category—is a common issue for AI and humans. Often it is addressed by computation of pairwise similarities between to-be-classified and already-classified objects. Similarity, however, can be computed in various ways. This project tests different mathematical similarities in classifiers.

You can read the scientific paper on PsychArXiv.