Interactive visual knowledge discovery from data-based temporal decision support system

Abstract

The article aims to present a generic interactive visual analytics solution that provides temporal decision support using knowledge discovery from data modules together with interactive visual representations. It bases its design decisions on classification of visual representation techniques according to the criteria of temporal data type, periodicity, and dimensionality. The design proposal is applied to an existing medical knowledge discovery from data–based decision support system aiming at assisting physicians in the fight against nosocomial infections in the intensive care units. Our solution is fully implemented and evaluated.

Explainable decision support through the learning and visualization of preferences from a formal ontology of antibiotic treatments

“…One of the perspectives opened up by this work is a comparison of the proposed visual tool with other possible presentation of the explanations, such as a simpler textual list of advantages and disadvantages, or more complex visual datamining tools such as the one proposed by H Ltifi et al [42] for monitoring nosocomial infections.…”

Visual Knowledge Generation from Data Mining Patterns for Decision-Making

“…The goal of this step is the translation of the patterns in a simple, natural and useful representation and description to the decision maker [15]. In this step we will rely on the principle of Shneiderman [20]to view the displayed information: (1) the representation of an overview of patterns, (2) extraction of points of interest and filtering irrelevant points.…”

“…This solution achieved positive results, but requires instead the integration of other tools to better attract the attention of users and then facilitate the decision-making. Recently, visualization techniques taking place in such systems provide interactive visual tools for more effective decision-making [15].…”

Combination of cognitive and HCI modeling for the design of KDD-based DSS used in dynamic situations

“…This dynamic process repeats itself during the patient hospitalization Tasks definition -Temporal data acquisition and preparation: in our database, the structure of the data is linear and generally based on lists. -Temporal data visualization: According to the critical choice of the temporal visualization technique proposed by [53], temporal data can be represented by the perspective wall technique [55] [56]. This technique allows joining together temporal and structural dimensions to present a global view (cf.…”