• Perceptions
• Other Applications

• Program
• Data Base

PERCEPTIONS - DATA BASE

Published DataSets in this web page can be downloaded and freely used.

In order to make use of it in your scientific publications the following three references must be included:

• Web page of research group SMIRE (http://bellman.ciencias.uniovi.es/SMIRE/perceptions.html) where the datasets to be analyzed, as well as the application perceptions providing it, can be found.
• González-Rodríguez, G.; Colubi, A.; Gil, M.A. (2012). Fuzzy data treated as functional data. A one-way ANOVA test approach. Computational Statistics and Data Analysis. Vol. 56(4), pp. 943–955. (doi:10.1016/j.csda.2010.06.013).
  In this paper, a guideline suggesting how to collect/describe fuzzy data associated with random experiments (and in particular for the perceptions experiment) is done. In addition a complete methodology for analyzing fuzzy data by using funtional data analysis techniques through Hilbert-valued random elements is also introduced.
• Colubi, A.; González Rodríguez, G.; Gil, M.A.; Trutschnig, W. (2011). Nonparametric criteria for supervised classification of fuzzy data. International Journal of Approximate Reasoning. Vol. 52(9), pp. 1272-1282. (doi:10.1016/j.ijar.2011.05.007).
  In this paper, some functional data analysis techniques are used in a supervised classification problem with fuzzy data.

Download Complete data set.

Download Data set in "Fuzzy data treated as functional data. A one-way ANOVA test approach".

Download Data set in "Nonparametric Criteria for Supervised Classification of Fuzzy Data".

DESCRIPTION OF THE DATA BASE

Different variables are separated by tab

Character strings are delimited by quotation marks " "

The name of each variable appears in the first row of the file

 

List of variables:

"EmailId" -> Identification index for the different emails (usually different individuals)

"sex" -> Sex of the individual coded as 0=Male, 1=Female

"nTrial" -> Number of trial (first 3 blocks of 9 observations, i.e. nTrial from 1 to 27, correspond to random ordered relative sizes in the fixed grid 3.7% 15.2% 26.8% 38.4% 50.0% 61.5% 73.1% 84.7% 96.3%, the remaining ones are just selected uniformly at random between 0 and 100)

"pHWS" -> Number of pixels in the horizontal direction of the Working Space

"Ref" -> Relative size of the reference line with respect to the available horizontal working space expressed in percentage (fixed to 80%)

"Pref" -> Absolute number of pixels (in horizontal sense) occupied by the reference line (i.e. integer part of Ref*HWS/100)

"Size" -> Number of pixels (in the horizontal direction) occupied by the line shown

"Rel.Size" -> Efective relative size of the line shown with respect to the reference line expressed in percentage (i.e. 100*Size/Pref). Note that depending on the screen resolution this relative size could be slightly different to the ones selected in the fixed grid for the first 27 trials.

"LingI" -> Linguistic description selected at first sign (0=Very Small, 1=Small, 2=Medium, 3=Large, 4=Very Large)

"LingF" -> Linguistic description selected after fixing the fuzzy response (0=Very Small, 1=Small, 2=Medium, 3=Large, 4=Very Large)

"Inf0" -> Infima of the 0-cut of the fuzzy set selected

"Inf1" -> Infima of the 1-cut of the fuzzy set selected

"Sup1" -> Suprema of the 1-cut of the fuzzy set selected

"Sup0" -> Suprema of the 0-cut of the fuzzy set selected

"Date" -> Date and time when the measurement was done

Note: EmailId 22 and 23 correspond to the same user (but using different email accounts)

HOW TO UPDATE THE DATA BASE

The data corresponding to your experiment is stored locally in a file named <your email address>.dat.
In order to access this data you have to send an email to Gil González-Rodríguez with the subject "Perceptions - Update Data Base" attaching this file.
The Data Base will be updated by adding your new information. You can continue the experiment afterwards so that sending the same file again means just an updating of the additional new information.

Back to top