Information Technology Reference
In-Depth Information
for J2ME and are built upon CLDC: KJava and
MIDP. Both KJava and MIDP are associated with
CLDC and smaller devices. A skeleton profile
upon which developers can create their profiles,
the Foundation Profile, is available for CDC
(“J2ME: Step by step”, see: http://www.digilife.
be/quickreferences/pt/).
system for medical diagnosis. It was reported
that the advanced system which used a fuzzified
dataset is more precise than the ordinary system
with continuous range of symptoms data. Also,
a series of optimizations on the multilayer per-
ceptron (MLP) classifier structure and training
epochs increased the system performance. The
best performance of 97.5% correct diagnosis
was achieved using fuzzified symptoms and an
optimized ANN with 15 hidden nodes and after
500 training epochs. An expert system to diagnose
plant diseases using a combination of descriptive
method and graphical representation was devel-
oped by Abu-Naser et al. (2008). A preliminary
evaluation of the system showed that the expert
system with the graphical representation is more
favourable than the descriptive one. Lopez et al.
(2009) developed a computer-aided diagnosis
system to assist clinicians in the Alzheimer's dis-
Related Work
Yana et al. (2006) developed a multilayer per-
ceptron-based medical decision support system
for the diagnosis of heart disease. The results
showed that multilayer perceptron-based decision
support system could achieve very high diagnosis
accuracy (>90%) and comparably small intervals
(<5%), proving its usefulness in support of clinic
decision process of heart diseases. Moein et al.
(2008) developed a hybrid fuzzy-neural automatic
Figure 3. J2ME Architecture (source: http://www.digilife.be/quickreferences/pt/)
Search WWH ::
Custom Search