1、1. 2. 3. 4. 5. 6. c-isodata7. 8. 1. 2. -3. 4. 5. 6. 5.1 -5.15.3()1 2 7. 13 4 k-lk-l- 5.15.1- 5.1 -11755.25.35.2 5.2 5.25.2.1 -5.3- 5.3k-l-k-l5.4 5.41 -ujuj2 3 uj4 5 ()()5.2.2 (5-1)(5-1)y1y2 (5-1)(supervised learning)(unsupervised learning)(supervised learning)(unsupervised learning) (supervised lear
2、ning)(unsupervised learning)knnsvm-(training data)bag-of-wordsk-meansk-means50-(gmm) classificationinductive learningdac/classification/prediction functionclassification modelprediction modelsupervised learning-unsupervised learning-knn3.1 3.2 3.3 -1. 2. 3. 4. 5. 6. (a) (b) -31. 2. 3. 3-13-13-1 3-1
3、3-1id3c4.5cart, 3.4 -naive bayestanbangbna, bpr(a)?0pr(b|a)?b pr(a)?0 pr(ab)?pr(b|a)pr(a)a1a2.an b(1) a1a2-an(2)?an?sb?an(3) pr(a)?0(n=1,2,-)n?1n?1?pr(ab)apr(a)pr(b)?pr(an)pr(b|an)n?1?ba1a2-an-pr(ai)-pr(b|ai)pr(b)a1a2-anb(1) a1a2-an(2)pra(nb)?prb()?an?1?n?sb?an(3) pr(a)?0(n=1,2,-)n?1pra(nb|?)pbr(an|(?prbn?1a|ja)p)nr( )par)j(ba1a2-an-pr(an)b-i-pr(ai|b)naive bayesnbda1,a2,-,anc|c|
