ARTÍCULO
TITULO

Does Removing/Replacing Missing Values Improve The Models' Classification Performances?

Jozef Zurada    

Resumen

The paper explores the effect of removing/replacing missing values on the classification performance of several models. The original data set, which contains a relatively large number of missing values, comes from the credit scoring context. This data set was not used to build the models, but it was converted to five other data sets with missing values either removed or replaced using different techniques. The models were built and tested on the five data sets. Preliminary computer simulation showed that the models created and tested on the four data sets in which missing values were replaced exhibited significantly better predictive performance than the model built and tested on the data set with missing values removed.

 Artículos similares

       
 
Fabrizio Stesina and Sabrina Corpino    
Given the role of Cubesats in the new space economy, a statistically relevant number of CubeSats have flown, and considering the high percentage of failed missions, the investigation of in-orbit anomalies becomes of paramount importance. It is rare to fi... ver más
Revista: Aerospace

 
Changgyun Kim, Youngdoo Son and Sekyoung Youm    
The aim of this study was to predict chronic diseases in individual patients using a character-recurrent neural network (Char-RNN), which is a deep learning model that treats data in each class as a word when a large portion of its input values is missin... ver más
Revista: Applied Sciences

 
Tobias Krueger     Pág. 138 - 148
Regulatory, low temporal resolution monitoring of freshwater quality does not fully capture the frequency distributions of the requisite parameters, particularly those that are highly skewed and heavy-tailed. Hence the summary statistics ultimately compa... ver más
Revista: Water Research

 
Maria Kovacova, Jan Balint     Pág. 89 - 98
Despite of expected growth of traffic from 1.75 % up to 2.4 % in period 2010 - 2050 in the airspace of Europe (Eurocontrol/STATFOR) there is still the need to ensure high level of safety performance. The main goal ? safe provision of Air Traffic Services... ver más

 
Walter Priesnitz Filho,Carlos Ribeiro,Thomas Zefferer     Pág. 81 - 96
Achieving interoperability, i.e. creating identity federations between different Electronic identities (eID) systems, has gained relevance throughout the past years. A serious problem of identity federations is the missing harmonization between various a... ver más