Levels_of_IVA Interactive_visual_analysis

1 levels of iva

1.1 base level
1.2 second level
1.3 third level
1.4 fourth level

levels of iva

the iva tools can divided several different levels of complexity. these levels provides user different interaction tools analyze data. uses, first level sufficient , level provides user fastest response interaction. higher levels make possible uncover more subtle relationships in data. however, requires more knowledge tools , interaction process has longer response time.

base level

the simple form of iva base level consists of brushing , linking. here user can set several views different dataset variables , mark interesting area in 1 of views. data points corresponding selection marked automatically in other views. lot of information can derived level of iva. datasets relationships between variables reasonably simple, technique sufficient user achieve required level of understanding.

second level

brushing , linking logical combination of brushes more advanced form of iva. makes possible user mark several areas in 1 or several views , combine these areas logical operators: and, or, not. makes possible explore deeper dataset , see more hidden information. simple example analysis of weather data: analyst might want discover regions both have warm temperatures , low precipitation.

third level

the logical combination of selections might not sufficient uncover meaningful information data set. there multiple techniques available make hidden relationships in data more apparent. 1 of these attribute derivation. allows user derive additional attributes data, such derivatives, clustering information or other statistic properties. in principle, operator can perform set of calculations on raw data. derived attributes can linked , brushed other attribute.

the second tool in level 3 of iva advanced brushing techniques, such angular brushing, similarity brushing or percentile brushing. these brushing tools select data points in more advanced fashion plain point , click selection. advanced brushing generates faster response attribute derivation, has higher learning curve , require deeper understanding of dataset.

fourth level

the fourth level of iva specific each dataset , varies dependent on dataset , purpose of analysis. calculated attribute specific data under consideration, belongs category. example analysis of flow data detection , categorization of vortexes or other structures present in flow data. means fourth-level iva techniques must individually tailored specific application. after detection of higher-order features, calculated attributes connected original data set , subjected normal technique of linking , brushing.