Database Reference
In-Depth Information
Characteristic set
Reference set
Any
Ordered
Has distances
Any
Even frequencies
of the values,
prevailing values,
rare values
Tendency toward
high, low or me-
dium values
Groups (clusters) of
references with
close characteristics
Linearly ordered
Constancy,
change, specific
value order
Increase, de-
crease, peak, low
point
Gradual change,
sharp change
Cyclically ordered
Frequency of
value appearing in
certain positions
of the cycle
Cyclical increase
and decrease
Gradual or sharp
changes within the
cycle and between
cycles
Has distances
Homogeneity or
heterogeneity,
large or small
regions of con-
gruency
Flatness, eleva-
tion, depression,
peak, depth, pla-
teau, valley
Smoothness (small
differences between
characteristics of
neighboring refer-
ences), abruptness
(big differences)
1.
For any kind of reference set: repeated pattern, frequent pattern, infrequent
pattern, prevailing pattern;
2.
For a linearly ordered reference set: specific sequence of patterns, alterna-
tion;
3.
For a cyclically ordered reference set: cyclically repeated pattern;
4.
For a reference set with distances: constant distance between repetitions of a
pattern, patterns occurring close to each other.
Any composite pattern may, in turn, be included in a bigger composite pattern, for
example, a frequently repeated pattern where increase is followed by decrease.
2.4
Directions of Further Work
What is presented in this section is only an initial sketch of the data-centered predic-
tive theory. Further work is required to ensure the comprehensiveness of the pattern
typology. Particular attention needs to be paid to multi-dimensional data. It is also
necessary to define pattern types used to represent relationships between attributes or
between phenomena (represented by several datasets differing in structure) such as
correlation (co-occurrence) or influence.
Then, it is necessary to evaluate Information Visualization techniques according to
the types of data they are suited for and the types of patterns they help to elicit. This
can form an appropriate basis for instructive topics and courses for users of Informa-
tion Visualization tools.
 
 
Search WWH ::




Custom Search