Fuzzy Set Theory Basic concept: Membership function: U, X is a subset of U, we define a membership function for X: e.g.: U = {a, b, c, d, e} X1 = {a/0.3, b/0.2, c/0.5} X2 = {a/0.2, b/0.1, d/0.5, e/0.2} X1 intersect X2 = {a/0.2, b/0.1} X1 union X2 = {a/0.3, b/0.2, c/0.5, d/0.5, e/0.2} ~X1(e) = 1 – muX1(e) = {a/0.7, b/0.8, c/0.5, d/1, e/1} X1 union ~X1 = {a/0.7, b/0.8, c/0.5, d/1, e/1} Fuzzy Relations: Binary fuzzy relations: Basic concept: U1 and U2 are universes, A binary relation R is a subset of U1 X U2 A fuzzy binary relation is determined by the membership function : : U1 X U2 [0, 1] Operations on fuzzy binary relations: 1. Projection: max operator 2. Extension: Cylindrical extension 3. Composition: max-min composition e.g.: X1 = {a/0.2, b/0,3, c/0.5} X2 = {a/0.3, b/0.3, c/0.4} R1 a a 0.3 b 0.3 c 0.5 b 0.3 0.3 0.5 c 0.4 0.4 0.5 Cylindrical extension R11 a a 0.5 b 0.5 c 0.5 b 0.5 0.5 0.5 c 0.5 0.5 0.5 Cylindrical extension R12 a a 0.4 b 0.4 c 0.5 b 0.4 0.4 0.5 c 0.4 0.4 0.5 b 0.4 0.4 0.5 c 0.4 0.4 0.5 b 0.5 0.5 0.5 c 0.5 0.5 0.5 R11 R12 a b c a 0.4 0.4 0.5 R1 (R11 R12) Cylindrical extension R11 a a 0.5 b 0.5 c 0.5 Cylindrical extension R12 a a 0.4 b 0.4 c 0.5 b 0.4 0.4 0.5 c 0.4 0.4 0.5 R11 R12 (Max-min Composition) a a 0.5 b 0.5 c 0.5 b 0.5 0.5 0.5 c 0.5 0.5 0.5 Fuzzy Rule-Based Systems: A: scores of projects [0, 100] B: final Scores A >= 90: [90, 100] B >= 90: [90, 100] If A >= 90 then B >= 90 Steps: 1. Identify variables V1, V2, …, Vm in an application 2. Determine domains Di’s of those variables (Universe of discourse) 3. Define fuzzy sets on Di’s: for each Di we associate a family of fuzzy sets, called reference fuzzy sets. e.g.: for a temperature variable T, we can define the fuzzy sets: Hot, Warm, Cool, Cold e.g.: Temp = [-100, 100] Fuzzification process: is a mapping of variable domains into fuzzy sets. Linguistic Variables: variables whose values are fuzzy sets. e.g.: the domain of T is the set {Hot, Warm, Cool, Cold}. Now, the value of T may be Hot or Cold, etc. Domain(Speed) = {Stop, Slow, Medium, Fast}. 4. Write fuzzy rules using the linguistic variables obtained from Step 3. e.g.: if T is Cold then Speed is Stop if T is Cool then Speed is Slow if T is Warm then Speed is Medium if T is Hot then Speed is Fast 5. Rule of inference: 6. Defuzzification Methods: Centroid or average e.g: X = {2,3,4,5,6,7,8,9} Y = {1,2,3,4,5,6} Medium = {0.1/2 + 0.3/3 + 0.7/4 + 1/5 + 1/6 + 0.7/7 + 0.5/8 + 0.2/9} Small = {1/1 + 1/2 + 0.9/3 + 0.6/4 + 0.3/5 + 0.1/6} If x is Medium then y is Small Use min Cartesian product, the rule is represented by the following relation R: 0.1 0.1 0.1 0.1 0.1 0.1 0.3 0.3 0.3 0.3 0.3 0.1 0.7 0.7 0.7 0.6 0.3 0.1 1 1 0.9 0.6 0.3 0.1 1 1 0.9 0.6 0.3 0.1 0.7 0.7 0.7 0.6 0.3 0.1 0.5 0.5 0.5 0.5 0.3 0.1 0.2 0.2 0.2 0.2 0.2 0.1 X is Small = {1/2 + 0.9/3 + 0.6/4 + 0.3/5 + 0.1/6} [1 0.9 0.6 0.3 0.1 0 0 0] o R = [0.6 0.6 0.6 0.6 0.3 0.1] (1*0.6 + 2 *0.6 + 3 * 0.6 + 4 * 0.6 + 5 * 0.3 + 6 * 0.1) / (0.6 + 0.6 + 0.6 + 0.6 + 0.3 + 0.1) = 8.1/2.8 = 2.89