Frequent Itemsets mining \u6700\u5e38\u89c1\u7684\u5f62\u5f0f\u5c31\u662f\u7ed9\u5b9amarket-basket\u5f62\u5f0f\u7684\u6570\u636e\uff08\u6bcf\u4e00\u884c\u76f8\u5f53\u4e8e\u4e00\u4e2a\u8d2d\u7269\u7bee\uff0c\u5305\u542b\u591a\u4e2a\u5546\u54c1item\uff09\uff0c\u7136\u540e\u6211\u4eec\u627e\u51fa\u5173\u8054\u5ea6\u5927\u4e8e\u67d0\u4e2a\u503c\u7684\u6240\u6709item\u96c6\u5408\u3002A-Priori\u7b97\u6cd5\u662fFrequent Itemsets mining\u91cc\u6700\u57fa\u672c\u7684\u7b97\u6cd5\u3002<\/p>\n
\u4e00\u3001A-Priori\u7684\u57fa\u672c\u601d\u8def<\/p>\n
\u7b2c\u4e00\u904d\uff08pass 1\uff09\uff0c\u626b\u63cf\u6587\u4ef6\uff0c\u7edf\u8ba1\u5355\u9879(single item)\u7684\u51fa\u73b0\u6b21\u6570\uff08\u4f7f\u75281\u4e2amap\u8fdb\u884c\u7edf\u8ba1\uff0ckey\u5c31\u662fitem\uff0cvalue\u5c31\u662f\u51fa\u73b0\u6b21\u6570\uff09\u3002\u6700\u540e\uff0c\u8fc7\u6ee4\u6389\u5c0f\u4e8e\u6700\u5c0f\u652f\u6301\u5ea6\u7684\uff0c\u5f97\u5230\u9891\u7e41\u5355\u9879\u96c6\u3002<\/p>\n
\u7b2c\u4e8c\u904d\uff08pass 2\uff09\uff0c\u626b\u63cf\u6587\u4ef6\uff0c\u5bf9\u4e8e\u6bcf\u4e00\u884c\uff0c\u5bf9\u4efb\u610f\u4e24\u4e2aitem\u7ec4\u5408\u5f97\u5230pair item\u3002\u5982\u679cpair item\u76842\u4e2a\u5355\u9879\u90fd\u5728\u9891\u7e41\u5355\u9879\u96c6\u91cc\uff0c\u5219\u7edf\u8ba1\u8fd9\u4e2apair item\u7684\u51fa\u73b0\u6b21\u6570\uff1b\u5426\u5219\u7565\u8fc7\u3002\u6700\u540e\u5f97\u5230\u6240\u6709pair item\u7684\u51fa\u73b0\u9891\u7387\uff0c\u8fc7\u6ee4\u6389\u5c0f\u4e8e\u6700\u5c0f\u652f\u6301\u5ea6\u7684\uff0c\u5f97\u5230\u9891\u7e412\u9879\u96c6\u3002<\/p>\n
\u7b2cN\u904d\uff0c\u5982\u679c\u53d1\u73b0\u9891\u7e41N-1\u9879\u96c6\u4e0d\u4e3a\u7a7a\uff0c\u5219\u8bf4\u660emining\u8fd8\u6ca1\u6709\u5b8c\u6210\uff0c\u9700\u8981\u8fdb\u884c\u7b2cn\u6b21\u626b\u63cf\u3002\u4ece\u6bcf\u884c\u4e2d\u53d6\u5f97\u4efb\u610fN\u4e2aitem\uff0c\u5982\u679c\u8fd9N\u9879\u7684\u6240\u6709N-1\u9879\u5b50\u96c6\u90fd\u5728\u9891\u7e41N-1\u9879\u96c6\u91cc\uff0c\u5219\u7edf\u8ba1\u5176\u51fa\u73b0\u6b21\u6570\uff0c\u5426\u5219\u7565\u8fc7\u3002\u6700\u540e\uff0c\u8fc7\u6ee4\u6389\u5c0f\u4e8e\u6700\u5c0f\u652f\u6301\u5ea6\u7684\uff0c\u5f97\u5230\u9891\u7e41N\u9879\u96c6\u3002<\/p>\n
\u76f4\u5230\u9891\u7e41\u9879\u96c6\u4e3a\u7a7a\u624d\u505c\u6b62\u5faa\u73af\u3002<\/p>\n
\u4e8c\u3001java\u5b9e\u73b0<\/p>\n
[code lang=”java”]
\nimport java.io.BufferedReader;
\nimport java.io.BufferedWriter;
\nimport java.io.FileNotFoundException;
\nimport java.io.FileOutputStream;
\nimport java.io.FileReader;
\nimport java.io.IOException;
\nimport java.io.OutputStreamWriter;
\nimport java.util.ArrayList;
\nimport java.util.HashMap;
\nimport java.util.HashSet;
\nimport java.util.Iterator;
\nimport java.util.LinkedHashMap;
\nimport java.util.List;
\nimport java.util.Map;
\nimport java.util.Set;
\nimport java.util.TreeSet;<\/p>\n
public class Apriori {<\/p>\n
\tprivate String inputFile;<\/p>\n
\tprivate double minSupport;<\/p>\n
\tprivate BufferedWriter bw;<\/p>\n
\tprivate HashSet<Set<Integer>> frequentItems = new HashSet<Set<Integer>>(); \/\/\u9891\u7e41\u9879\u96c6<\/p>\n
\tprivate Set<Integer> frequentSingleItems = new HashSet<Integer>(); \/\/\u9891\u7e41\u5355\u9879\u96c6<\/p>\n
\tprivate int totalCount; \/\/\u9879\u76ee\u6570\u91cf<\/p>\n
\tpublic Apriori(String inputFile, double minSupport) {
\n\t\tthis.inputFile = inputFile;
\n\t\tthis.minSupport = minSupport;
\n\t}<\/p>\n
\t\/**
\n\t * \u627e\u51fa\u9891\u7e41\u4e00\u9879\u96c6
\n\t * @return
\n\t * @throws IOException
\n\t *\/
\n\tpublic Map<Set<Integer>,Integer> findF1Item() throws IOException {
\n\t\tMap<Set<Integer>,Integer> result = new LinkedHashMap<Set<Integer>, Integer>();
\n\t\tMap<Integer, Integer> map = new HashMap<Integer, Integer>();
\n\t\tBufferedReader reader = new BufferedReader(new FileReader(inputFile));
\n\t\tString line;
\n\t\tint numberOfLine = 0;
\n\t\twhile ((line = reader.readLine()) != null) {
\n\t\t\tnumberOfLine++;
\n\t\t\tString[] items = line.split(" ");
\n\t\t\tfor(String item : items){
\n\t\t\t\tint intItem = Integer.valueOf(item);
\n\t\t\t\tif (map.containsKey(intItem)) {
\n\t\t\t\t\tmap.put(intItem, map.get(intItem) + 1);
\n\t\t\t\t} else {
\n\t\t\t\t\tmap.put(intItem, 1);
\n\t\t\t\t}
\n\t\t\t}
\n\t\t}
\n\t\ttotalCount = numberOfLine;
\n\t\treader.close();
\n\t\t\/\/\u4f7f\u7528TreeSet\u6309\u7167item\u7f16\u53f7\u4ece\u5c0f\u5230\u5927\u6392\u5e8f
\n\t\tTreeSet<Integer> treeSet = new TreeSet<Integer>(map.keySet());
\n\t\tfor (Integer item : treeSet) {
\n\t\t\tint count = map.get(item);
\n\t\t\t\/\/\u8fc7\u6ee4\u6389\u51fa\u73b0\u9891\u7387\u5c0f\u4e8e\u6700\u5c0f\u652f\u6301\u5ea6\u7684item
\n\t\t\tif (count >= minSupport*totalCount) {
\n\t\t\t\tSet<Integer> f1Set = new TreeSet<Integer>();
\n\t\t\t\tf1Set.add(item);
\n\t\t\t\tresult.put(f1Set, count);
\n\t\t\t\tfrequentItems.add(f1Set);
\n\t\t\t\tfrequentSingleItems.add(item);
\n\t\t\t}
\n\t\t}
\n\t\treturn result;
\n\t}<\/p>\n
\tpublic Map<Set<Integer>,Integer> generateNextPass(int k) throws Exception{
\n\t\tMap<Set<Integer>, Integer> map = new HashMap<Set<Integer>, Integer>();
\n\t\tBufferedReader reader = new BufferedReader(new FileReader(inputFile));
\n\t\tString line;
\n\t\twhile ((line = reader.readLine()) != null) {
\n\t\t\tString[] items = line.split(" ");
\n\t\t\tList<Set<Integer>> list = generateSubset(items, k);
\n\t\t\tfor(Set<Integer> set :list){
\n\t\t\t\tif(map.containsKey(set)){
\n\t\t\t\t\tmap.put(set, map.get(set)+1);
\n\t\t\t\t}
\n\t\t\t\telse{
\n\t\t\t\t\tmap.put(set, 1);
\n\t\t\t\t}
\n\t\t\t}
\n\t\t}
\n\t\treader.close();
\n\t\tfrequentItems = new HashSet<Set<Integer>>();
\n\t\tfrequentSingleItems = new HashSet<Integer>();
\n\t\tIterator<Set<Integer>> ite = map.keySet().iterator();
\n\t\twhile(ite.hasNext()){
\n\t\t\tSet<Integer> key = ite.next();
\n\t\t\tint value = map.get(key);
\n\t\t\tif(value<totalCount*minSupport){
\n\t\t\t\tite.remove();
\n\t\t\t}
\n\t\t\telse{
\n\t\t\t\tfrequentItems.add(key);
\n\t\t\t\tfor(int item: key){
\n\t\t\t\t\tfrequentSingleItems.add(item);
\n\t\t\t\t}
\n\t\t\t}
\n\t\t}
\n\t\treturn map;
\n\t}<\/p>\n
\tprivate List<Set<Integer>> generateSubset(String[] array, int k) throws Exception{
\n\t\tList<Set<Integer>> result = new ArrayList<Set<Integer>>();
\n\t\tint[] intArray = new int[array.length];
\n\t\tfor(int i=0; i<array.length;i++){
\n\t\t\tintArray[i] = Integer.parseInt(array[i]);
\n\t\t}
\n\t\tint[] newArray = filterItems(intArray);
\n\t\tList<Set<Integer>> list = generateSubSets(newArray,k);
\n\t\tfor(Set<Integer> set: list){
\n\t\t\t\/\/\u5c06set\u53d8\u6210\u6570\u7ec4
\n\t\t\tint smallArray[] = new int[set.size()];
\n\t\t\tint i=0;
\n\t\t\tIterator<Integer> ite = set.iterator();
\n\t\t\twhile(ite.hasNext()){
\n\t\t\t\tsmallArray[i]= ite.next();
\n\t\t\t\ti++;
\n\t\t\t}
\n\t\t\t\/\/\u627e\u51faset\u7684\u6240\u6709k-1\u6b21subItemSet
\n\t\t\tList<Set<Integer>> smallList = generateSubSets(smallArray, k-1);
\n\t\t\t\/\/\u5982\u679c\u67091\u4e2asubItemSet\u4e0d\u662f\u9891\u7e41\u7684\uff0c\u5219\u5224\u65adset\u4e0d\u662f\u9891\u7e41\u7684
\n\t\t\tboolean flag = true;
\n\t\t\tfor(Set<Integer> item: smallList){
\n\t\t\t\tif(!frequentItems.contains(item)){
\n\t\t\t\t\tflag = false;
\n\t\t\t\t\tbreak;
\n\t\t\t\t}
\n\t\t\t}
\n\t\t\tif(flag){
\n\t\t\t\tresult.add(set);
\n\t\t\t}
\n\t\t}
\n\t\treturn result;
\n\t}<\/p>\n
\tpublic void printFrequentItems(Map<Set<Integer>,Integer> itemSets, int i) throws FileNotFoundException, IOException {
\n\t\tif(bw == null){
\n\t\t\tbw = new BufferedWriter(new OutputStreamWriter(new FileOutputStream("output.txt")));
\n\t\t}
\n\t\tStringBuffer sb = new StringBuffer();
\n\t\tfor (Set<Integer> set : itemSets.keySet()) {
\n\t\t\tsb.append("[");
\n\t\t\tfor (Integer str : set) {
\n\t\t\t\tsb.append(str + " ");
\n\t\t\t}
\n\t\t\tsb.append("], count:");
\n\t\t\t\t\tsb.append(itemSets.get(set));
\n\t\t\t\t\tsb.append("\\n");
\n\t\t}
\n\t\tbw.write(sb.toString());
\n\t\tbw.flush();
\n\t}<\/p>\n
\t\/**
\n\t* \u8fc7\u6ee4\u6389\u4e0d\u5728\u524d\u4e00\u6b21\u9891\u7e41\u5355\u9879\u96c6\u91cc\u7684item
\n\t* @param array
\n\t* @return
\n\t*\/
\n\tprivate int[] filterItems(int[] array){
\n\t\tList<Integer> list = new ArrayList<Integer>();
\n\t\tfor(int i=0;i<array.length;i++){
\n\t\t\tif(frequentSingleItems.contains(array[i])){
\n\t\t\t\tlist.add(array[i]);
\n\t\t\t}
\n\t\t}
\n\t\tint[] newArray = new int[list.size()];
\n\t\tfor(int i=0;i<newArray.length;i++){
\n\t\t\tnewArray[i] = list.get(i);
\n\t\t}
\n\t\treturn newArray;
\n\t}<\/p>\n
\tpublic void closeOutputWriter() throws IOException{
\n\t\tif(bw == null){
\n\t\t\tbw.close();
\n\t\t}
\n\t}<\/p>\n
\tprivate List<Set<Integer>> generateSubSets(int[] array, int k){
\n\t\tList<Set<Integer>> list = new ArrayList<Set<Integer>>();
\n\t\tif(array.length<k){
\n\t\t\treturn list;
\n\t\t}
\n\t\t\/\/ \u521d\u59cb\u5316\u79fb\u4f4d\u6cd5\u9700\u8981\u7684\u6570\u7ec4
\n\t\tbyte[] bits = new byte[array.length];
\n\t\tfor (int i = 0; i < bits.length; i++) {
\n\t\t\tbits[i] = i < k ? (byte) 1 : (byte) 0;
\n\t\t}
\n\t\tboolean find = false;
\n\t\tdo {
\n\t\t\t\/\/ \u627e\u523010\uff0c\u6362\u621001
\n\t\t\tSet<Integer> set = getCombination(array, bits);
\n\t\t\tif(set!=null && set.size()!=0){
\n\t\t\t\tlist.add(set);
\n\t\t\t}
\n\t\t\tfind = false;
\n\t\t\tfor (int i = 0; i < array.length – 1; i++) {
\n\t\t\t\tif (bits[i] == 1 && bits[i+1] == 0) {
\n\t\t\t\t\tfind = true;
\n\t\t\t\t\tbits[i] = 0;
\n\t\t\t\t\tbits[i+1] = 1;
\n\t\t\t\t\tif(bits[0] == 0){
\n\t\t\t\t\t\tfor (int p=0, q=0; p < i; p++){
\n\t\t\t\t\t\t\tif(bits[p] == 1){
\n\t\t\t\t\t\t\t\tbyte temp = bits[p];
\n\t\t\t\t\t\t\t\tbits[p] = bits[q];
\n\t\t\t\t\t\t\t\tbits[q] = temp;
\n\t\t\t\t\t\t\t\tq++;
\n\t\t\t\t\t\t\t}
\n\t\t\t\t\t\t}
\n\t\t\t\t\t}
\n\t\t\t\t\tbreak;
\n\t\t\t\t}
\n\t\t\t}<\/p>\n
\t\t} while (find);
\n\t\treturn list;
\n\t}<\/p>\n
\tprivate Set<Integer> getCombination(int[] array, byte[] bits) {
\n\t\tSet<Integer> set = new TreeSet<Integer>();
\n\t\tfor (int i = 0; i < bits.length; i++) {
\n\t\t\tif (bits[i] == (byte) 1) {
\n\t\t\t\tset.add(array[i]);
\n\t\t\t}
\n\t\t}
\n\t\treturn set;
\n\t}<\/p>\n
}<\/p>\n
[\/code]<\/p>\n
[code lang=”java”]<\/p>\n
import java.util.Map;
\nimport java.util.Set;<\/p>\n
public class Main {<\/p>\n
\tpublic static void main(String[] args) {
\n\t\tSystem.out.println("program starts\u2026");
\n\t\tlong startTime = System.currentTimeMillis();
\n\t\tString inputFile = "src\/test.txt";
\n\t\tdouble minSupport = 0.02;
\n\t\tApriori apriori = new Apriori(inputFile, minSupport);
\n\t\ttry {
\n\t\t\tSystem.out.println("pass 1");
\n\t\t\tMap<Set<Integer>, Integer> f1Set = apriori.findF1Item();
\n\t\t\tapriori.printFrequentItems(f1Set, 1);
\n\t\t\tMap<Set<Integer>, Integer> result = f1Set;
\n\t\t\tint i = 2;
\n\t\t\tdo {
\n\t\t\t\tSystem.out.println("pass " + i);
\n\t\t\t\tresult = apriori.generateNextPass(i);
\n\t\t\t\tapriori.printFrequentItems(result, i);
\n\t\t\t\ti++;
\n\t\t\t} while (result.size() != 0);
\n\t\t\tapriori.closeOutputWriter();
\n\t\t} catch (Exception e) {
\n\t\t\te.printStackTrace();
\n\t\t}
\n\t\tlong endTime = System.currentTimeMillis();
\n\t\tSystem.out.println("execution time:" + (endTime – startTime) + "ms");
\n\t}
\n}<\/p>\n
[\/code]<\/p>\n
\u8f93\u5165\u6587\u4ef6\uff1atest.txt<\/p>\n
1 2 5
\n2 4
\n2 3
\n1 2 4
\n1 3
\n2 3
\n1 3
\n1 2 3 5
\n1 2 3<\/p>\n
\u8f93\u51fa\u6587\u4ef6\uff1aoutput.txt<\/p>\n
[1 ], count:6
\n[2 ], count:7
\n[3 ], count:6
\n[4 ], count:2
\n[5 ], count:2
\n[1 2 ], count:4
\n[1 3 ], count:4
\n[1 4 ], count:1
\n[2 3 ], count:4
\n[2 4 ], count:2
\n[1 5 ], count:2
\n[2 5 ], count:2
\n[3 5 ], count:1
\n[1 2 3 ], count:2
\n[1 2 4 ], count:1
\n[1 2 5 ], count:2
\n[1 3 5 ], count:1
\n[2 3 5 ], count:1
\n[1 2 3 5 ], count:1<\/p>\n","protected":false},"excerpt":{"rendered":"
Frequent Itemsets mining \u6700\u5e38\u89c1\u7684\u5f62\u5f0f\u5c31\u662f\u7ed9\u5b9amarket-basket\u5f62\u5f0f\u7684\u6570\u636e\uff08\u6bcf […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[20,19],"class_list":["post-56","post","type-post","status-publish","format-standard","hentry","category-8","tag-a-priori","tag-19"],"_links":{"self":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts\/56","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=56"}],"version-history":[{"count":8,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts\/56\/revisions"}],"predecessor-version":[{"id":133,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts\/56\/revisions\/133"}],"wp:attachment":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=56"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=56"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=56"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}