Apriori算法實驗報告(共7頁)

1、河北大學數學與計算機學院實 驗 報 告課程名稱 數據挖掘 實驗項目 編程實現(xiàn)Apriori算法 實驗儀器 VC+6.0 專 業(yè)_計算機軟件與理論_ 學 號 20101180 學生姓名 喬紅光 實驗日期 2011年4月9日 實驗地點 實驗室 成 績 指導教師 袁方 一、實驗目的1.加強對Apriori算法的理解；2.鍛煉分析問題、解決問題并動手實踐的能力；二、實驗內容編程實現(xiàn)Apriori算法；三、實驗要求 編程實現(xiàn)Apriori算法，加深對其理解。四、實驗準備1.看懂Apriori算法的基本思想；2.上網查閱相關資料。五、實驗過程主要代碼如下：void CAprioriView:Apriori

2、Gen(int nCandFreqItem, int nMinSupp)/由L(k-1)生成C(k)CString strTemp1,strTemp2,strRightTemp1,strRightTemp2;CString strTemp,strLeftTemp1,strLeftTemp2;int nstrTemp1,nstrTemp2; int nCandFreqItemCount=0;strTemp1=;strTemp2=;strRightTemp1=;strRightTemp2=;strTemp=;strLeftTemp1=; strLeftTemp2=;nstrTemp1=0;nstr

3、Temp2=0;nAllFreqItem=nAllFreqItem + LargeItemCountnCandFreqItem-1;for(int i1=0;i1LargeItemCountnCandFreqItem-1;i1+) strTemp1=LargeItemnCandFreqItem -1i1; nstrTemp1=strTemp1.ReverseFind(,);strRightTemp1=strTemp1.Right(strTemp1.GetLength ()-nstrTemp1-1); strLeftTemp1=strTemp1.Left(nstrTemp1);for(int j

4、1=i1+1;j1LargeItemCountnCandFreqItem-1;)/strTemp2=strCandFreqItemj1;strTemp2=LargeItemnCandFreqItem-1j1;nstrTemp2=strTemp2.ReverseFind (,);strRightTemp2=strTemp2.Right(strTemp2.GetLength ()-nstrTemp2-1);strLeftTemp2=strTemp2.Left(nstrTemp2) ;if(strLeftTemp1=strLeftTemp2)&(strRightTemp1strRightTemp2)

5、strTemp=strTemp1+,+strRightTemp2;if(Prune(nCandFreqItem,strTemp)CandLargeItemnCandFreqItemnCandFreqItemCount+=strTemp;j1+;elsej1+;else break;CandLargeItemCountnCandFreqItem=nCandFreqItemCount;void CAprioriView:SubItemGen(int strSubItemCount,CString strSubItem)/對每個事務分解成單個項目CString strTemp1;CString st

6、rTempSubItem10;CString strReverse;int nSubItemCount;int nstrRightTemp1;int nTempCount;strTemp1=strSubItem;nSubItemCount=0;while(nstrRightTemp1=strTemp1.ReverseFind(,)!=-1)nTempCount=strTemp1.GetLength() - nstrRightTemp1 - 1; strTempSubItemnSubItemCount+=strTemp1.Right( nTempCount);strTemp1=strTemp1.

7、Left(nstrRightTemp1); strTempSubItemnSubItemCount+=strTemp1;for(int i2=0;i2nSubItemCount/2;i2+)strReverse=strTempSubItemnSubItemCount-i2-1;strTempSubItemnSubItemCount- i2-1=strTempSubItemi2;strTempSubItemi2=strReverse; for(int i1=0;i1GetODBCFieldCount();int nCount=0;int nFreqItem100; int nDbCount=0;

8、CString strInit;CString strValue;CString strIntToString; / 初始化候選項目集 for(int nInitCount=0;nInitCountnItemCount;nInitCount+) strInit=;strInit.Format(%s%d,strInit,nInitCount+1); CandLargeItem0nInitCount=I+strInit;/初始化數組for(int ii=0;iiMoveFirst ();while(!m_pSet-IsEOF()for(int j=1;jGetFieldValue(j,strVal

9、ue); SubItemGen(nDbCount+ ,strValue);for(int i=0;iMoveNext();nDbItemCount=nDbCount; for(int i1=0;i1=dItemSupp) LargeItem0nCount=CandLargeItem0i1; m_List_FreqItem.InsertItem(nCount,strIntToString);m_List_FreqItem.SetItemText(nCount,0,LargeItem0nCount); strIntToString=;strIntToString.Format(%s%d,strIn

10、tToString,nFreqItemi1);m_List_FreqItem.SetItemText(nCount,1,strIntToString);nCount+;LargeItemCount0=nCount;BOOL CAprioriView:Prune(int nCandFreqItemCount,CString strCandFreqItem)CString strTemp1;CString strTempSubItemnMaxSize;CString strReverse;CString strSubCandFreqItemnMaxSize;/分解候選項目CString strSu

11、bTemp=;CString strSubTemp1;int nSubFreqItemCount=0;/統(tǒng)計分解候選項目的個數int nSubItemCount;int nstrRightTemp1;int nTempCount;int nPruneCount=0;strTemp1=strCandFreqItem;nSubItemCount=0;while(nstrRightTemp1=strTemp1.ReverseFind(,)!=-1)nTempCount=strTemp1.GetLength() - nstrRightTemp1 - 1; strTempSubItemnSubItemC

12、ount+=strTemp1.Right( nTempCount);strTemp1=strTemp1.Left(nstrRightTemp1); strTempSubItemnSubItemCount+=strTemp1;for(int i2=0;i2=0;i3-) strSubTemp1=strTempSubItemi3;strSubTemp.Empty(); for(int i4=0;i4nSubItemCount;i4+)if(strSubTemp1!=strTempSubItemi4) strSubTemp=strSubTemp+strTempSubItemi4+,;strSubCa

13、ndFreqItemnSubFreqItemCount+ = strSubTemp.Left(strSubTemp.GetLength()-1); for(int i5=0;i5nSubFreqItemCount;i5+) for(int i6=0;i6=0) nPruneCount+;if(nPruneCount=nSubItemCount)return TRUE;return FALSE;void CAprioriView:ShowFreqItem(int nScanCount)CString strIntToString=; CString strValue;CString strjj3

14、2;int nLargeCount=-1;int nLargeItemCount=0; /以下為求頻繁項目集int k,nListFreqItemCount;k=nScanCount; nListFreqItemCount=LargeItemCountk-1;m_List_FreqItem.InsertItem(0,strValue);m_List_FreqItem.SetItemText(0,0,-); m_List_FreqItem.SetItemText(0,1,-); for(int jj2=0;jj2=dItemSupp)LargeItemknLargeItemCount+=Cand

15、LargeItemkjj2;nLargeCount+;strjj31=strIntToString;strjj30=CandLargeItemkjj2;strIntToString=;strIntToString.Format(%s%d,strIntToString,nCountCandjj2); strjj31=strIntToString;m_List_FreqItem.InsertItem(nLargeCount,strValue);m_List_FreqItem.SetItemText(nLargeCount,0,LargeItemknLargeItemCount-1);m_List_FreqItem.SetItemText(nLargeCount,1,strIntToString);UpdateWindow(); /復制頻繁項目個數 LargeItemCountk=nLargeItemCount;運行結果圖如下：六、實驗總結