If it was so, it might be; and if it were so, it would be; but as it isn't, it ain't. That's logic. -- Lewis Carroll

:: CLUE Lab ::


Computer Languages and User Experience Laboratory

Università degli Studi di Salerno
Dipartimento di Informatica
 

Visual Languages

A visual language is a set of practices by which images can be used to communicate concepts.

In the ambit of Visual Languages, our research has focused on theoretical, methodological and applied aspects of the traditional visual languages and has deepened on the study of the sketched visual languages, that is, the languages defined on free-hand drawn sketched visual languages.

As for the theoretical aspects, extensions to the model of picturel languages to the colored, drawn and pixel pictures case have been defined, studying some of their computability and decidability aspects.
As for the methodological and implementing aspects, a technique, based on positional extended grammars, for the creation of the compilers and the visual languages and then of the visual modeling environment through the use of technologies tradiotionally employed for the textual programming languages, has been define. Some applications focus on the use of visual languages both in the e-learning and in the software engineering fields.

In the ambit of sketched visual languages, our research has brought to the definition of a new grammar formalism, called Sketch Grammar, capable of describing both the free hand drawn symbols and the syntax of the languages they compose. On the sketch grammars several procedures have been defined, similar to those defined for LR Parsing, for the automatic generation of recognizers for the described sketch languages.
In order to improve the efficiency and the accuracy of the built recognizers, several strategies have been defined, such as, the introduction of a training phase, the use of a framework based on agents for the low level recognition of symbols, the use of the ambiguity resolution context during symbol recognition, the use of techniques for error recovery to capture information on missing or erroneously recognized strokes.
The results gained in this ambit and further studies on the definition of incremental parsers based on GLR parsing have been used for the automatic generation of syntax-aware visual language editors, which allow the user to easily construct a hand drawn visual phrase. In these editors, the user is interactively aided in the correct building of the phrase with suggestions on the possible extensions of the phrase with techniques of automatic completion of incomplete symbols.
An initial experimental prototype has been integrated in the VLDesk system, already used for the generation of classical visual environments. This has allow the development of empirical studies on real languages, such as UML, to verify the effectiveness of the proposed approaches.

 

Publications

Our recent publications in Visual Languages research

2010

2009

2008

2007

2006

2005

2004

CLUE Lab Site Search

Upcoming Events

No current events.