中科院数据挖掘第一次作业(最新2016年).docx

1.2.3.使用了matlab和c语言编程，代码附在后面。(a)meanmedianstdAge46.4444 5113.2186%fat28.783330.70009.2544(b)Fig.1 age boxplotFig.2 %fat boxplot(c)Fig.3 scatter plot(d)000.110.110.420.470.630.680.710.760.820.820.870.890.920.920.9710.050.5400.290.680.520.560.560.670.7710.610.740.650.760.720.960.8(e)correlation_coefficient = 1.0000 0.81760.8176 1.0000 显著的正相关。(f)#include stdio.hint main(int argc, char *argv[]){double sum,avg,fat[18]={7.8,9.5,17.8,25.9,26.5,27.2,27.4,28.8,30.2,31.2,31.4,32.9,33.4,34.1,34.6,35.7,41.2,42.5};for(int i=0;i18;i++){sum+=fat[i];if((i+1)%6==0){avg=sum/6;sum=0;printf(%f\n,avg);}}return 0;}箱子1：19.116667,19.116667,19.116667,19.116667,19.116667,19.116667箱子2：30.316667,30.316667,30.316667,30.316667,30.316667,30.316667箱子3：36.916667,36.916667,36.916667,36.916667,36.916667,36.916667,(g)箱子1：7.8,7.8,27.2,27.2,27.2,27.2箱子2：27.4,27.4,32.9,32.9,32.9,32.9箱子3：33.4,33.4,33.4,33.4,42.5,42.5% boxplot([y1; y2; y3].)boxplot(age);boxplot(fat);%meanf=mean(fat)%meana=mean(age)%ma=median(age)%mf=median(fat)%stda=std(age)%stdf=std(fat)%scatter(age,fat)% % % % % Normalize% min_age=min(age);% max_age=max(age);% for i=1:length(age)% new_age(i)=(age(i)-min_age)/(max_age-min_age);% end% new_age% new_age=roundn(new_age,-2)% min_fat=min(fat);% max_fat=max(fat);% for i=1:length(fat)% new_fat(i)=(fat(i)-min_fat)/(max_fat-min_fat);% end% new_fat=roundn(new_fat,-2)% % correlation_coefficient=corrcoef(age,fat)% % % % % bin means% sort_fat=sort(fat)#include stdio.hint main(int argc, char *argv[]){double sum,avg,fat[18]={7.8,9.5,17.8,25.9,26.5,27.2,27.4,28.8,30.2,31.2,31.4,32.9,33.4,34.1,34.6,35.7,41.2,42.5};for(int i=0;i18;i++){if((fat[i]-fat[(i/6)*6])(fat[(i/6)*6+5]-fat[i]))fat[i]=fat[(i/6)*6+5];elsefat[i]=fat[(i/6)*6];printf(%3f\n,fat[i]);}return 0;}4.5.(a) 绝对支持度min_sup=4*60%=2.4。第一次扫描，得到频繁项集合支持度计数按递减顺序排序，记为L，L={{A:4},{B:4},{C:3}}构造FP树如下：FP树只包含单路径，故频繁模式集为{{A},{B},{C},{A,B},{A,C},{B,C},{A,B,C}}。(b) 对于该数据库，Apriori算法需要扫描3次数据库，而FP-Growth只需扫描2次，随后的操作都在内存中进行。并且FP-Growth使用最不频繁的项做后缀，提供了