Decision Theoretic Fusion Framework for Actionability Using Data Mining on an Embedded System - Data Mining: Theory, Methodology, Techniques, and Applications - page 92

The content of this paper is as follows. The design concept of the proposed sys-
tem is presented in Section 2. In Section 3, we describe the data mining method-
ologies based on the decision theoretic fusion framework for actionability. Finally,
in Section 4, we provide discussions and conclusive remarks.
2 Architecture of Embedded Car Navigation System
This Section describes the introduction of embedded system on a real car for Telemat-
ics service. In addition, speech interactive agent is described as a effective speech
interaction tools for safeguard driving and service guide as well as information gath-
ering and generation tools.
2.1 System Overview
The embedded car navigation system provides the various embedded service applica-
tions on a car as well as networked service applications via a wireless network using
the CDMA 2000 terminal as shown in Figure 1. In our proposed system, we include
the interactive techniques using speech interactive agent to provide a speech interac-
tion method as an intelligent interface between human and machine. The speech
interactive agent plays a role in combining and processing the information from inter-
face modalities as well as in communicating with the data mining server to provide
useful information to the user. This system needs a database to store some valuable
information and manage some information. Such an embedded system has limited
resources. To resolve this problem, this system stores and manages the multiple
Fig. 1. System architecture overview