ARTIFACTS / Software / Card Sorting Visualization tool
In the context of the PhiloSurfical project, a card-sorting experiment was organized in order to gather empirical evidence regarding philosophers' implicit conceptualizations of the domain. Card-sorting is a specific type of knowledge acquisition experiment, in which experts are given a set of cards representing important domain concepts and are asked to repeatedly sort them according to known criteria.
In order to analyze the data we made use of a purposely-built software tool that allows the visualization of multiple perspectives on the data. This is an ontology-based web-application which supports the navigation of the data and the automatic creation of html tables aimed at facilitating the interpretation of the results. For example by clicking on a respondent name it is possible to see all of his/her sorts and categories in one table, while, by clicking on a meta-criteria we can see which criteria and categories it comprises, etc..
The application relies on a simple ontology with five classes (namely volunteer, criterion, category, card and meta-criteria), which has been instantiated using the data from the experiment. The application uses OCML for the knowledge modeling and data-storage, the lisp programming language for the server and back-end functionalities, html and ajax techniques for the user interface.
Although it has been implemented only as an ad-hoc solution for card sorting data visualization, the application turned out to be of fundamental importance for the success of the experiment. Hence we do not exclude that with some extra effort it could be developed into a generic framework for supporting the visualization of data elicited through card-sorting (get in touch for more details.)
In order to analyze the data we made use of a purposely-built software tool that allows the visualization of multiple perspectives on the data. This is an ontology-based web-application which supports the navigation of the data and the automatic creation of html tables aimed at facilitating the interpretation of the results. For example by clicking on a respondent name it is possible to see all of his/her sorts and categories in one table, while, by clicking on a meta-criteria we can see which criteria and categories it comprises, etc..
The application relies on a simple ontology with five classes (namely volunteer, criterion, category, card and meta-criteria), which has been instantiated using the data from the experiment. The application uses OCML for the knowledge modeling and data-storage, the lisp programming language for the server and back-end functionalities, html and ajax techniques for the user interface.
Although it has been implemented only as an ad-hoc solution for card sorting data visualization, the application turned out to be of fundamental importance for the success of the experiment. Hence we do not exclude that with some extra effort it could be developed into a generic framework for supporting the visualization of data elicited through card-sorting (get in touch for more details.)
