演示文稿(第十次课) 机器学习 教学课件.pptVIP

* Choose initial “seed” events from E Determine a star for each seed against the other seed events By appropriately modifying and selecting complexes from stars, construct a disjoint cover of E that optimizes the criterion LEF Is the termination Criterion satisfied? END a Given: E-a set of data events k-the number of clusters LEF-the clustering quality criterion Y N Is the clustering quality improving? Choose k new seeds which Are central events Choose k new seeds which are “border” events a Y N 2 1 0 2 1 0 2 1 0 e10 e9 c e8 e7 b e6 a e5 c b e4 a e3 c b a e1 e2 X1 X2 0 1 2 X4 X3 0 1 2 2 2 c 2 e10 0 0 c 2 e9 2 1 b 2 e8 1 0 b 2 e7 0 1 a 2 e6 1 1 c 1 e5 2 0 a 1 e4 2 1 c 0 e3 0 0 b 0 e2 1 0 a 0 e1 X4 X3 X2 X1 Event d f a b c K=2 ; LEF-sparseness, Complexity; Termination criterion:base=2,probe=2 Iteration 1 Step 1: Select seed: e1,e2 Step 2: Produce Stars: RG(e1|e2,m) RG(e2|e1,m) m=5 RG(e1|e2,m)={[x2=a][x3=0∨1],[X4=1 ∨2]} RG(e2|e1,m)={[x2=b ∨c],[x4=0 ∨2]} Generalize: RG(e1|e2,m)={[x2=a][x3≦1],[X4=1∨2]} RG(e2|e1,m)={[x2=f],[x4=0∨2]} Step 3: Evaluation and Modification(disjoint) Sparseness Complexity Complex 1: [x2=a][x3≦1] 15 2 Complex 2: [x2=f] 47 1 62 3 (b) Complex 1: [x4=1∨2] Complex 2: [X2=f] (c) Complex 1: [x2=a][x3≦1] Complex 2: [X4=0∨2] (d) Complex 1: [x4=1∨2] Complex 2: [x4=0∨2] Step 4: The termination criterion is tested Step 5:select new seeds {e1,e4,e6} {e2,e3,e5,e7,e8,e9,e10} Central events: e4,e8 Iteration 2 Step 2: Produce satrs RG(e4|e8,m ), RG(e8|e4,m) RG(e4|e8,m)={[x2=a][x3≦1],[x1≦1][x3 ≦1],[x3=0]} RG(e8|e4,m)={[x1=2],[x2=f],[x3≧1]}