数据挖掘实验报告Word格式.docx
《数据挖掘实验报告Word格式.docx》由会员分享，可在线阅读，更多相关《数据挖掘实验报告Word格式.docx（19页珍藏版）》请在冰点文库上搜索。
匕surr#Jci归忙之址
humiiit^zxhijhp-nonrriar
分类规则：
J48prunedtreeoutlook=sunny
|humidity=high:
no（3.0）
|humidity=normal:
yes（2.0）outlook=overcast:
yes（4.0）outlook=rainy
|windy=TRUE:
no（2.0）
|windy=FALSE:
yes（3.0）剪枝后结果为
QWeteExplorer|=|^||S3|
|Pt>
e■->
n.jd^SS-!
|iZliqisL»
r|&
.w■.«
c|qt•|£
J■匚k■.!
trIbq4■■&
|Vs■«
事■
SIi.tLli
分类精度变为57.1%性能变好
（1）C4.5算法优缺点
优点：
分类精度高，生成的分类规则比较简单，易于理解。
缺点：
需要多次扫描数据集，比较低效
五、分析与讨论
六、教师评语
签名：
成绩
日期：
实验项目名称：
KNh算法
一、实验目的及要求
1:
KNN算法的基本思路、步骤。
选择UCI中的5个标准数据集，使用KNN算法在该数据集上计算混淆矩阵。
选择2个数据集，选择不同的k值，k=1,3,5,7,9，对比KNN算法计算结果的差异。
二、实验设备（环境）及要求
四、实验内容与步骤
1.数据集contact-lenses.aff
Glass.aff
两者的混淆矩阵分别为
==Confus'
iQnMate!
暮=
abc＜一一cla-ssifiedas
2121a-safi:
121Ib=hard
4011I□•none
Matrix
===ConfusionMatrix===
a
b
c
d
e
f
g
<
--
classifiedas
55
9
6
01
=buildwindfloat
1551
4
3
2
11
b=
=buildwindnon-float
5
c:
=vehicwindfloat
=vehicwindnon-float
010
1|
e=
=containers
1
7
f=tableware
23|
g=
=headlamps
（2）两个数据集在K=1,3,5,7,9下结果分别为
Glass：
K=1;
===Summary===
===DetailedAccuracyByClass===
K=3;
CorrectlyClassifiedInstances
154
71.9626%
IncorrectlyClassifiedInstances
60
28.0374%
Kappastatistic
0.6097
0.0983
Meanabsoluteerror
Rootmeansquarederror0.2524
Relativeabsoluteerror46.4438%
Rootrelativesquarederror77.7792%
TotalNumberofInstances214
5910
a:
1954
b:
10
8
2|
223|
g:
K=5;
67.757%
32.243%
CorrectlyClassifiedInstances145
IncorrectlyClassifiedInstances69
Kappastatistic0.5469
Meanabsoluteerror0.1085
Rootmeansquarederror0.2563
Relativeabsoluteerror51.243%
Rootrelativesquarederror78.9576%
TPRateFPRatePrecisionRecallF-MeasureROCAreaClass
0.843
0.229
0.641
0.728
0.867
buildwindfloat
0.6840.174
buildwindnon-float
0.684
0.848
0.010
0.642
vehicwindfloat
00
?
vehicwindnon-float
0.385
0.025
0.5
0.435
0.952
containers
0.667
0.01
0.75
0.706
0.909
tableware
0.793
0.016
0.885
0.836
0.89
headlamps
WeightedAvg.0.6780.1420.6350.6780.6510.853
abcdefg<
--classifiedas
591010000|a=buildwindfloat
205210300|b=buildwindnon-float12500000|c=vehicwindfloat0000000|d=vehicwindnon-float0500503|e=containers
0200160|f=tableware
12001223|g=headlamps
K=7;
===Summary===
64.0187%
35.9813%
CorrectlyClassifiedInstances137
IncorrectlyClassifiedInstances77
Kappastatistic0.4948
Meanabsoluteerror0.1147
Rootmeansquarederror0.2557
Relativeabsoluteerror54.1689%
Rootrelativesquarederror78.7876%
0.829
0.271
0.598
0.695
0.876
0.6050.181
0.648
0.605
0.626
0.852
0.059
0.005
0.105
0.71
0.308
0.03
0.4
0.348
0.939
0.556
0.015
0.625
0.588
0.976
WeightedAvg.0.640.1580.6360.640.6170.864
581110000|a=buildwindfloat
264600400|b=buildwindnon-float
11510000|c=vehicwindfloat
0000000|d=vehicwindnon-float
0500413|e=containers
1200150|f=tableware
K=9;
135
63.0841%
79
36.9159%
Kappastatistic（
0.4782
Meanabsoluteerror
0.1196
Rootmeansquarederror
0.2581
Relativeabsoluteerror
56.4924%
Rootrelativesquarederror
79.5178%
TotalNumberofInstances
214
0.278
0.592
0.69
0.881
0.645
0.174
0.671
0.658
0.853
0.694
0（
0.231
0.333
0.273
0.933
0.222
0.286
0.964
0.027
0.821
0.807
0.888
WeightedAvg.
0.631
0.159
0.58
0.6310.597
FPRate
PrecisionRecall
0.864
581110000|a=buildwindfloat
234900310|b=buildwindnon-float
13
0|
d=
4|
f=
contact-lenses:
19
79.1667%
20.8333%
0.6262
0.2262
0.3165
59.8856%
72.4707%
24
===DetailedAccuracyByClass
===
TPRateFPRate
F-Measure
ROCArea
0.8
0.053
0.80.8
0.958
0.1
0.60.75
0.925
0.8570.8
0.828
0.896
WeightedAvg.0.792
0.167
0.8020.792
0.795
0.914
Classsofthardnone
=soft
=hard
212|
c=
:
none
CorrectlyClassifiedInstances1979.1667%
IncorrectlyClassifiedInstances520.8333%
Kappastatistic0.6262
Meanabsoluteerror0.2262
Rootmeansquarederror0.3165
Relativeabsoluteerror59.8856%
Rootrelativesquarederror72.4707%
TotalNumberofInstances24
soft
0.6
hard
0.857
0.792
0.802
TPRateFPRatePrecision
RecallF-Measure
Class
abc<
401|a=soft
031|b=hard
1212|c=none
66.6667%
33.3333%
CorrectlyClassifiedInstances16
IncorrectlyClassifiedInstances8Kappastatistic0.3356
0.2793
Rootmeansquarederror
0.3624
RelativeabsoluteerrorRootrelativesquarederrorTotalNumberofInstances
73.9227%
82.9705%
TPRateFPRatePrecisionRecallF-MeasureClass
0.947
0.25
0.856
0.859
0.375
0.653
0.655
0.877
302|a=soft
013|b=hard
CorrectlyClassifiedInstances14IncorrectlyClassifiedInstances10
Kappastatistic-0.0619
Meanabsoluteerror0.3188
Rootmeansquarederror0.387
Relativeabsoluteerror84.3959%
Rootrelativesquarederror88.61%
58.3333%
41.6667%
PrecisionRecallF-Measure
00.053000
00000
0.831
0.9331
0.6090.9330.7370.807
WeightedAvg.0.5830.6360.38
0.5830.461
0.841
005|a=soft
004|b=hard
1014|c=none
TotalNumberofInstances24===DetailedAccuracyByClass===
TPRateFPRatePrecisionRecallF-MeasureROCArea
0.609
0.7370.807
Weighte
升级会员 