Information Technology Reference
In-Depth Information
abstracting and transforming for multi-aspect analysis and simulation. The
data dimension supports the implementation of a grid-based, analysis and
simulation oriented, dynamic, spatial and multimedia database for storing
and sharing the heterogeneous brain data eciently and effectively.
- The experiment dimension is a conceptual model of domain knowledge
aiming at the systematic experimental design in BI methodology. It de-
scribes characteristics of various experimentation plans, their classification
and inter-relationships at the conceptual level. Systematic experimental de-
sign is an important issue of BI methodology. For uncovering the principles
and mechanisms of HIPS, BI researchers need to design a series of cognitive
experiments to obtain high quality of experimental data, which represent
different aspects of various thinking centric cognitive functions, based on
systematic methodology of cognitive experimental design. Thus, Data-Brain
needs to include an experiment dimension. By relations with the function di-
mension and data dimension, the experiment dimension explicitly describes
various relationships among various data sources. By the experiment dimen-
sion, cognitive experiments related information can be stamped on each data
set for supporting systematic data analysis and simulation.
- The analysis dimension is a conceptual model of domain knowledge aim-
ing at the systematic data analysis and simulation in BI methodology. It de-
scribes characteristics of various analysis and simulation methods, as well as
their relationships with multiple human brain data for multi-aspect analysis
and simulation. Agent-enriched data mining for multi-aspect data analysis is
an important issue of BI methodology because the brain is too complex for
a single data mining algorithm to analyze all the available data. Thus, Data-
Brain needs to include an analysis dimension. Based on the analysis dimen-
sion corresponding to the data and experiment dimensions, various methods
for data processing, mining, reasoning, and simulation can be deployed as
agents on a multi-phase process for performing multi-aspect analysis as well
as multi-level conceptual abstraction and learning, which aims at discovering
useful knowledge to understand human intelligence in depth [12].
As shown in Fig. 1, these dimensions are not isolated from each other, but with
various relations among them.
4
Data-Brain Modeling
4.1
Data-Brain Modeling Based on Brain Informatics Methodology
As a network/grid level of conceptual data model, Data-Brain modeling can
adopt an ontological modeling approach. Comparing with other ontology engi-
neering methodologies, the “Enterprise Methodology” [8] especially focuses on
the construction of concept hierarchy and more fit to construct a Data-Brain
which take concept hierarchies as skeletons of dimensions. Thus, we choose it to
construct a Data-Brain. The course mainly includes the following four steps:
Search WWH ::




Custom Search