The Exposition of Fuzzy Decision Trees and Their Application in Biology - Advancing Artificial Intelligence Through Biological Process Applications - page 392

problems that may have inherent uncertainties as
those described.
There are a number of future issues that are
associated with the general decision tree methodol-
ogy which are pertinent with the work in biological
applications. These include the generality of the
results accrued from the FDT analyses, which
technically moves the issue to the complexity
of the FDT constructed. This complexity issue
includes the, number, size and structure, of the
membership functions used to fuzzify the data
considered. This is a common issue with all tech-
niques that operate within a fuzzy environment.
Further, with respect to FDTs, there is also the
role of pruning which refers to the cutting back
of the established paths in a FDT.
Perhaps the future trends for the impact of
FDTs in biology based applications is the abil-
ity of the biology orientated analysts to acquire
the required level of understanding surrounding
FDTs that make their employment practical and
pertinent.

REFERENCES

Armand, S., Watelain, E., Roux, E., Mercier,
M., & Lepoutre, F.-X. (2007). Linking clinical
measurements and kinematic gait patterns of
toe-walking using fuzzy decision trees. Gait &
Posture , 25 (3), 475-484.

Beynon, M. J., Buchanan, K., & Tang, Y.-C.
(2004a). The Application of a Fuzzy Rule Based
System in an Exposition of the Antecedents of
Sedge Warbler Song Flight. Expert Systems ,
21 (1), 1-10.

Beynon, M.J., Peel, M.J., & Tang, Y.-C. (2004b).
The Application of Fuzzy Decision Tree Analysis
in an Exposition of the Antecedents of Audit Fees.
OMEGA - International Journal of Management
Science , 32 (2), 231-244.

Breiman, L. (2001). Statistical Modeling: The Two
Cultures. Statistical Science , 16 (3), 199-231.

Chang, R.L.P., & Pavlidis, T. (1977). Fuzzy deci-
sion tree algorithms. IEEE Transactions Systems
Man and Cybernetics , SMC-7(1), 28-35.

CONCLUSION

Chiang, I-J., & Hsu, J.Y-J. (2002). Fuzzy clas-
sification trees for data analysis. Fuzzy Sets and
Systems , 130 , 87-99.

The interpretative power of fuzzy decision trees
(FDTs) and their ability to handle uncertain in-
formation, may make it an appropriate technique
to be employed in biological application analysis.
That is, biologists are regularly investigating the
physiology/relationship of objects (animals etc.)
with their environment. The interpretation issue
outlined, due to the readability of fuzzy decision
rules constructed, could change the emphasis of
what is wanted from an analysis.
Within classification problems, the standard
notion of performance of a technique, like mul-
tivariate discriminant analysis, is classification
accuracy. However, their interpretative power
may be limited. The FDT analyses outlined does
allow for both interpretation and accuracy.
The material presented in this chapter has
shown FDTs with emphasis on the role of read-
ability in the results found.

Daan, S., Barnes, B.M., & Strijkstra, A.M. (1991).
Warming up for sleep? Ground squirrels sleep
during arousals from hibernation. Neuroscience
Letters , 128 , 265-268.

Dombi, J., & Gera, Z. (2005). The approximation
of piecewise linear membership functions and
Łukasiewicz operators. Fuzzy Sets and Systems ,
154 , 275-286.

Garibaldi, J.M., & John, R.I. (2003). Choosing
Membership Functions of Linguistic Terms,
Proceedings of the 2003 IEEE International
Conference on Fuzzy Systems , St. Louis, USA,
pp. 578-583.