基于決策樹的西瓜分類辨別

學(xué)院信息與電專業(yè)_計(jì)算機(jī)科學(xué)與技術(shù)班級(jí)1502本學(xué)號(hào) 密封學(xué)學(xué)院信息與電專業(yè)_計(jì)算機(jī)科學(xué)與技術(shù)班級(jí)1502本學(xué)號(hào) 密封學(xué)生須將文字寫在此 任課教 成（可指定題目，也可說明題目范圍基于決策樹的西類辨要求（ 題目、內(nèi)容、行文、字?jǐn)?shù)等作出判分規(guī)定1、結(jié)合自己感的實(shí)際應(yīng)用領(lǐng)域，自擬題目，并使用本學(xué)數(shù)據(jù)挖掘課程中所學(xué)習(xí)的基本數(shù)據(jù)分析和處理方法，包括數(shù)據(jù)預(yù)探索性分析數(shù)據(jù)倉(cāng)庫(kù)相似度計(jì)算決策樹分類最近鄰分類樸素分類、邏輯回歸、支持向量機(jī)、神經(jīng)網(wǎng)絡(luò)、關(guān)聯(lián)規(guī)則、means聚類層次聚類和密度聚類等知識(shí)對(duì)所解決的實(shí)際問題從本理論數(shù)學(xué)原理算法設(shè)計(jì)數(shù)據(jù)描述和實(shí)驗(yàn)結(jié)果與分析等角度開描述2、要 結(jié)構(gòu)清晰、語言通順、圖文并茂，字?jǐn)?shù)不少于正文首行縮進(jìn)2個(gè)漢字多倍行距設(shè)置為1.25，字體要求漢字宋體，英文為TimesNewRoman，大小為小四。 寫自己動(dòng)手設(shè)計(jì)算法和編寫程序如發(fā) 來自網(wǎng)上或他人， 處理，本門課程考核總成績(jī)計(jì)0分 版和程序文件夾；文件夾 名格式統(tǒng)一為“學(xué)號(hào)-名-班級(jí) 題目此外必須上 的打印稿并要與電子稿內(nèi)容一致，不規(guī)范者將酌情扣分教師評(píng)教師簽 日一.IF-THEN規(guī)則的預(yù)測(cè)樹。1、構(gòu)建決策樹階段：使用ID3（ItiveDichotomizer3，迭代二分器3）算4二數(shù)學(xué)原理Entropy(t)p(j|t)logp(j|j其中：p(j|t)tjk

色澤,根蒂,敲聲,紋理,臍部,觸感,0-1000100000001000200000010011110111110010111110-022021-222220-200221-010100-211100-110011-200220-001110-0,1,0,1,0,1,0,1,GAINsplitEntropy(p)nEntropy(i) pk個(gè)劃分時(shí)，nii個(gè)劃分中的記錄數(shù)三.

0,1,0,0,否-usingnamespacestd;#definenaturen7#definetestnconstintINF=(1<<30)-1;typedefstructnode{intstructnode*child[3];////vector<structnode*>}node,constintmaxn=boolvisi[maxn];//標(biāo)記屬性的數(shù)組,1constintmaxt=100;boolvisj[maxt];//inttestdata[maxn][maxt];doublegain[maxn];

intnaturechildnum[7]={0,3,3,3,3,3,2//stringname[8]是","色澤","根蒂","敲聲","紋理","臍部","觸感","否voidinit_testdata(){cout輸入訓(xùn)練集endl;for(intj=0;j<testn;j++){for(inti=1;i<=naturen;{cin>>}}}//intget_nature(int&fchild,int{doubledoublecount0=0,count1=0,counts=0;for(intj=0;j<testn;j++){if(visj[j]=={//if(testdata[j][naturen]=={}}}//cout<<count0<<count1<<counts<<endl;doubleen1=0,en2=0;if(count0!={en1=-(count0/counts)*log(count0/}if(count1!={en2=-(count1/counts)*log(count1/}entropy=intfnature=-1;doublemaxgain=-doublecountchild0=0,countchild1=0,countchild2=0,countchilds=0;doublecountchild0_b0=0,countchild0_b1=0;

doublecountchild1_b0=0,countchild1_b1=0;doublecountchild2_b0=0,countchild2_b1=for(inti=1;i<=naturen;{if(visi[i]=={for(intj=0;j<testn;{if(testdata[j][i]=={//if(testdata[j][naturen]=={}//if(testdata[j][naturen]==-{}}if(testdata[j][i]=={//if(testdata[j][naturen]=={}//if(testdata[j][naturen]==-{

}

}

if(countchild0_b1!={}

}

if(testdata[j][i]=={//if(testdata[j][naturen]=={}//if(testdata[j][naturen]==-{}}

//e1=-(countchild0/countchilds)*(p0+}if(countchild1!={doublep0=0,p1=0;if(countchild1_b0!=0){ }if(countchild1_b1!={ }doublee1=0,e2=0,e3=0;if(countchild0!=0){doublep0=0,p1=0;if(countchild0_b0!=0){

//e2=-(countchild1/countchilds)*(p0+}if(countchild2!={doublep0=0,p1=}

if(countchild2_b0!={ }if(countchild2_b1!={ }//e3=-(countchild2/countchilds)*(p0+

if(countchild1=={fchild=behavior=}if(countchild1=={fchild=behavior=-}if(countchild2=={fchild=gain[i]=entropy-e1-e2-if{maxgain=gain[i];fnature=i;if(countchild0=={fchild=behavior=}if(countchild0=={fchild=behavior=-}

behavior=}if(countchild2=={fchild=behavior=-}}}}visi[fnaturetrue;for(intj=0;j<testn;{if(testdata[j][fnature]=={visj[j]=}}returnfnature;}//voidcreate_d_tree(pnode{if(curnode=={

if(curnode->data==-1||curnode->data=={}for(inti=0;i<naturechildnum[curnode->data];{}}//curnode=(pnode)malloc(sizeof(node));intfchild=-1,ben=-2;curnode->data=get_nature(fchild,for(inti=0;i<naturechildnum[curnode->data];{curnode->child[i]==}//childif(fchild!=-{pnodecurnodechild=(pnode)malloc(sizeof(node));curnodechild->data=ben;for(inti=0;i<3;{curnodechild->child[i]==}curnode->child[fchild]=}

voidtest_out(pnode{if(root==NULL)if(root->data!=-cout<<name[root->data]<<cout<<name[7]<<endl;for(inti=0;i<3;i++){}}int{fill(visi,visi+maxn,0);fill(visj,visj+maxt,pnoderoot;root=(pnode)malloc(sizeof(node));intfchild=-1,beh=-2;root->dataget_nature(fchild,beh);////childif(fchild!=-{pnodeye(pnode)malloc(sizeof(node));ye->data=beh;//獲取屬性for(inti=0;i<3;{ye->child[i]==}root->child[fchild]=}return0;}四.0,色澤,根蒂,敲聲,紋理,臍部,觸感,密度,含糖率,1,青綠,蜷縮,濁響,清晰,凹陷,硬滑,0.697,0.46,2,烏黑,蜷縮,沉悶,清晰,凹陷,硬滑,0.774,0.376,3,烏黑,蜷縮,濁響,清晰,凹陷,硬滑,0.634,0.264,4,青綠,蜷縮,沉悶,清晰,凹陷,硬滑,0.608,0.318,

5,淺白,蜷縮,濁響,清晰,凹陷,硬滑,0.556,0.215,6,青綠,稍蜷,濁響,清晰,稍凹,軟粘,0.403,0.237,7,烏黑,稍蜷,濁響,稍糊,稍凹,軟粘,0.481,0.149,9,烏黑,稍蜷,沉悶,稍糊,稍凹,硬滑,0.666,0.091,10,青綠,硬挺,清脆,清晰,平坦,軟粘,0.243,0.267,12,淺白,蜷縮,濁響,模糊,平坦,軟粘,0.343,0.099,13,青綠,稍蜷,濁響,稍糊,凹陷,硬滑,0.639,0.161,14,淺白,稍蜷,沉悶,稍糊,凹陷,硬滑,0.657,0.198,15,烏黑,稍蜷,濁響,清晰,稍凹,軟粘,0.36,0.37,16,淺白,蜷縮,濁響,模糊,平坦,硬滑,0.593,0.042,17,青綠,蜷縮,沉悶,稍糊,稍凹,硬滑,0.719,0.103, 色澤,根蒂,敲聲,紋理,臍部,觸感,0-1000100000001000200000010011110111110010111110-022021-222220-200221-010100-211100-110011-200220-001110-0,1,0,1,0,1,0,1,0,1,0,0,否-//constintINF=(1<<30)-//typedefstruc