K-DIME: a software framework for Kansei filtering of Internet material

Abstract

The paper presents K-DIME, a system for retrieving material from the Internet on the basis of both the objective and subjective (Kansei) features of the content. It is implemented in CAPE and uses DIME, a set of software components for identifying and coordinating the use of external knowledge sources able to identify relevant items and assess their Kansei properties appropriately for the user. It is designed to allow external knowledge sources-both image databases and Kansei "oracles"-to be added easily.