Weighted and unweighted distances based decision tree for ranking data

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Preference data represent a particular type of ranking data (widely usedin sports, web search, social sciences), where a group of people gives their preferencesover a set of alternatives. Within this framework, distance-based decisiontrees represent a non-parametric tool for identifying the profiles of subjects givinga similar ranking. This paper aims at detecting, in the framework of (completeand incomplete) ranking data, the impact of the differently structured weighted distancesfor building decision trees. The traditional metrics between rankings don’ttake into account the importance of swapping elements similar among them (elementweights) or elements belonging to the top (or to the bottom) of an ordering(position weights). By means of simulations, using weighted distances to build decisiontrees, we will compute the impact of different weighting structures both onsplitting and on consensus ranking. The distances that will be used satisfy Kemenysaxioms and, accordingly, a modified version of the rank correlation coefficient τx,proposed by Edmond and Mason, will be proposed and used for assessing the trees’goodness.
Original languageEnglish
Title of host publicationBook of short Papers SIS 2018
Pages1-7
Number of pages7
Publication statusPublished - 2018

Fingerprint

Dive into the research topics of 'Weighted and unweighted distances based decision tree for ranking data'. Together they form a unique fingerprint.

Cite this