A visual sonificated web search clustering engine

Information Visualization is a widespread approach of the Information and Communication Technologies, and it facilitates the manipulation of abstract information by representing it by geometric models. It has been shown that spatial representation can be independent by the sensorial way in which it is perceived (Avraamides et al. 2004; De Vega et al. 2001), leading to the hypothesis of an amodal spatial representation (Bryant 1992). In light of these studies, an important alternative to Information Visualization methods appears to be the Sonification approach, an information representation method (Olivetti Belardinelli et al. 2009) that implements non-speech audio information to ‘‘represent data relations into perceived relations in an acoustic signal for the purposes of facilitating communication and interpretation’’ (Kramer et al. 1997). The main focus of the sonification approach is on the interactivity between the user and the data representation, since it allows user to manipulate complex data and their inner relations in a dynamic way. Currently, the information representation techniques of main search engines are based on a top-down hierarchic order, according to quantitative indexes of ranking. In a recent study (Borsci et al. 2008; Federici et al. 2009), it has been shown that high ranking values of web popularity (WP) do not match with high levels of accessibility. The top-down output of search engines leads users to assign a qualitative value to the hierarchy organization based on WP, although it is calculated without considering quality indexes (i.e., accessibility). Moreover, Andronico et al. highlighted the lack of qualitative ratings on WP, showing that the main search engines do not comply with W3C accessibility guidelines (Andronico et al. 2008). In this study, we aim to describe the redesign process of WhatsOnWeb (WoW) (Di Giacomo et al. 2007, 2008), a visual graph based search engine implemented by the Department of Electronic and Information Engineering of the University of Perugia. WoW is an autonomous application able to order indexed Web data using semantic nodes in a single page network diagram. In this way, WoW overcomes the page scrolling normally required by the reports of the traditional search engines (Search Engines Report Page), overpassing the limits of traditional Web information representation methods pointed out by Borsci et al. (2008).