METHODOLOGYPhases/steps carried out in developing WES doinclude: 1.Problem definition/Scope identification. 2.Knowledge acquisition. 3. Knowledge representation.4. Coding. 5. Testing and implementation.
3.1 Problem Definition/Scope Identification.
An expert system needs precise domain. Thedomain must be well organized and well understood.In diagnosis domain, as number of disorders(diseases) increase linearly, the number of possiblediagnoses increase exponentially (i.e. there are morecombinations of diagnoses to consider). This type of growth in the total no. of solutions is calledcombinatorial explosion. Clearly, Skin Disordersdiagnoses are an appropriate domain for expert systemdevelopment. This WES has a narrow domain of skindiseases. Following list shows some sample skindiseases which can be diagnosed by our system.I.
Acne -> Acne vulgaris, Acne Scars, AcneTreatment, Back Acne, Baby Acne, AcneMyths, Adult Acne, Acne Control, TeaTree oil for AcneIX.
3.2 Knowledge Acquisition
Knowledge Acquisition includes three activities:choosing what knowledge is needed; obtaining andinterpreting the knowledge; and modeling theknowledge. Some commonly used approaches aredirect interviews, observations Ishikawa diagrams,case studies . Direct interviews and observationswere used for KA. The KA for this system consistedof several interviews with skin specialists, makingobservations and getting historical data from variousskin clinics, depts. and wards in DHQ teachinghospitals of Pakistan , free skin caring camps and skinlabs of medical college. Knowledge acquisitionprocess will last for three months.
3.3 Knowledge Representation
There are numerous approaches for knowledgerepresentation depending on nature of problemdomain. As this is a rule based system, so IF-THENstyle rules are used because they are easy tounderstand and enhance. The rule has twocomponents: IF<situation THEN<suggestion> TheIF-part (antecedent/left hand side) suggests describesa situation under which the rule is applicable. TheTHEN part (consequent/right hand side) suggests anaction to be performed under the situation ( for actionrules) or a plausible inference to make under thesituation( for inference rules).
3.4 Testing and implementation
System will be tested to ensure that it providesgood quality decisions. Reasoning technique is testedfor correctness i.e. whether it attains the desiredaccuracy rate. During testing omitted rules, incorrectrules, overlooked special cases are checked.Association of rules and missing relationships are alsochecked. Thorough testing verifies that all paths in theknowledge base are covered. ES was tested by twoskin Specialists. The prototyping approach will beused to implement WES. A small domain of theknowledge acquired will be implemented in JESS andpresented to the skin specialist. They will test theprototype using different scenarios. Theirrecommendations, additions, deletions or changesfrom the conclusions given by WES. Once the skinspecialists agreed with the recommendations given bythe ES, new prototype with more knowledge will bedeveloped and presented again to experts for testing.The procedure will continue until all the acquiredknowledge was included in WES knowledge base.WES is planned to be launched on the web usingApache server. Working Model of our expert systemis given as under in Figure 1.
Figure 1. Working Model of our Expert System
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 6, June 2011324http://sites.google.com/site/ijcsis/ISSN 1947-5500