Fuzzy Expert System For Diabetes Using Fuzzy Verdict Mechanism – Article Example

FUZZY EXPERT SYSTEM FOR DIABETES USING FUZZY VERDICT MECHANISM LECTURER DATE
Introduction
Diabetes has continually increased globally at alarming numbers. This has captured the attention of biometrical engineering and artificial intelligence. As a result, technologies such as fuzzy expert systems have been incorporated to enhance a more advanced diabetes research. This system gives a detailed description for diabetes and supports the validation of medical practitioners.
Architecture of fuzzy expert systems
The following is the architecture for fuzzy expert systems application.
The main objective of fuzzy concept is to reassign the PIDD information into the knowledge that is required. The numbers in the fuzzy system should be constructed in accordance to the concepts that are generated. Each case in PIDD has nine characteristics. Each of these characteristics can be constructed as a fuzzy variable in compliance with some fuzzy numbers. the characteristics are given in form of a table. The following is a table of the nine PIDD features( Lumpur1.)
Acronym
Full name
elements
Glucose
Concentration of plasma glucose in two hours
mg/dl
Pregnant
The number of times pregnant
-
INS
Serum insulin- two hours
Mu
u/ml
DBP
Diastolic blood pressure
mmhg
TFST
Triceps skin fold thickness
mm
BMI
Body mass index
Kg/m2
Age
age
-
DM
Diabetes mellitus where I is explained as tested positive for diabetes
-
Fuzzification
This is a procedure of changing a crisp input value into a degree as required by the terms. In case of a raise as a result of ambiguity or vagueness, the variable is probably fuzzy. It can only be signified by a membership function (Lumpur2). If the source f inputs is from a piece of hardware, are a drive from sensor measurement, then the inputs in numerical form can be fuzzified so that they can be used in the fuzzy system.
The verdict mechanism assumes the possibility of a person testing positive for diabetes for each instance in fuzzification. It then conveys these possibilities in form of sentences. This verdict mechanism evaluates the personal physical data, and then translates the outcome into knowledge and presents the results inform f of descriptions. The descriptions include a weak, strong or very strong case in diabetes.
Conclusion
Results gathered from detailed experiments propose that the fuzzy expert system can process data and convey the information acquired into knowledge to simulate the thinking process of a person. Further, the results suggest that the system works more effectively for applications on diabetes and previous cases of development in diabetes.
Work cited
Informatics engineering and information science International Conference, ICIEIS 2011, Kuala Lumpur, Malaysia, November 14-16, 2011, proceedings. Berlin New York: Springer, 2011.