mapreduce实现"浏览该商品的人大多数还浏览了"经典应用
输入:
日期 ...cookie id. ...商品id..
xx xx xx
输出:
商品id 商品id列表(按优先级排序,用逗号分隔)
xx xx
比如:
id1 id3,id0,id4,id2
id2 id0,id5
整个计算过程分为4步
1、提取原始日志日期,cookie id,商品id信息，按天计算,最后输出数据格式
商品id-0 商品id-1
xx x x
这一步做了次优化,商品id-0一定比商品id-1小，为了减少存储，在最后汇总数据转置下即可
reduce做局部排序及排重
2、基于上次的结果做汇总,按天计算
商品id-0 商品id-1 关联值(关联值即同时访问这两个商品的用户数)
xx x x xx
3、汇总最近三个月数据,同时考虑时间衰减,时间越久关联值的贡献越低,最后输出两两商品的关联值（包括转置后)
4、行列转换，生成最后要的推荐结果数据,按关联值排序生成
第一个MR
import java.io.IOException;import java.util.ArrayList;import org.apache.hadoop.conf.Configuration;import org.apache.hadoop.fs.FileSystem;import org.apache.hadoop.fs.Path;import org.apache.hadoop.io.LongWritable;import org.apache.hadoop.io.Text;import org.apache.hadoop.io.WritableComparable;import org.apache.hadoop.io.WritableComparator;import org.apache.hadoop.mapreduce.Job;import org.apache.hadoop.mapreduce.Mapper;import org.apache.hadoop.mapreduce.Partitioner;import org.apache.hadoop.mapreduce.Reducer;import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;import org.apache.hadoop.util.GenericOptionsParser;import org.apache.log4j.Logger;/* * 输入:原始数据,会有重复 *日期 cookie 楼盘id * * 输出: * 日期 楼盘id1 楼盘id2 //楼盘id1一定小于楼盘id2 ,按日期 cookie进行分组 * */public class HouseMergeAndSplit {public static class Partitioner1 extends Partitioner<TextPair, Text> { @Override public int getPartition(TextPair key, Text value, int numParititon) { return Math.abs((new Text(key.getFirst().toString()+key.getSecond().toString())).hashCode() * 127) % numParititon; }} public static class Comp1 extends WritableComparator { public Comp1() { super(TextPair.class, true); } @SuppressWarnings("unchecked") public int compare(WritableComparable a, WritableComparable b) { TextPair t1 = (TextPair) a; TextPair t2 = (TextPair) b; int comp= t1.getFirst().compareTo(t2.getFirst()); if (comp!=0) return comp; return t1.getSecond().compareTo(t2.getSecond()); }} public static class TokenizerMapper extends Mapper<LongWritable, Text, TextPair, Text>{ Text val=new Text("test"); public void map(LongWritable key, Text value, Context context ) throws IOException, InterruptedException { String s[]=value.toString().split("\001"); TextPair tp=new TextPair(s[0],s[1],s[4]+s[3]); //thedate cookie city+houseid context.write(tp, val); } } public static class IntSumReducer extends Reducer<TextPair,Text,Text,Text> { private static String comparedColumn[] = new String[3]; ArrayList<String> houselist= new ArrayList<String>(); private static Text keyv = new Text(); private static Text valuev = new Text(); static Logger logger = Logger.getLogger(HouseMergeAndSplit.class.getName()); public void reduce(TextPair key, Iterable<Text> values, Context context ) throws IOException, InterruptedException { houselist.clear(); String thedate=key.getFirst().toString(); String cookie=key.getSecond().toString(); for (int i=0;i<3;i++) comparedColumn[i]=""; //first+second为分组键,每次不同重新调用reduce函数 for (Text val:values) { if (thedate.equals(comparedColumn[0]) && cookie.equals(comparedColumn[1])&& !key.getThree().toString().equals(comparedColumn[2])) { // context.write(new Text(key.getFirst()+" "+key.getSecond().toString()), new Text(key.getThree().toString()+" first"+ " "+comparedColumn[0]+" "+comparedColumn[1]+" "+comparedColumn[2])); houselist.add(key.getThree().toString()); comparedColumn[0]=key.getFirst().toString(); comparedColumn[1]=key.getSecond().toString(); comparedColumn[2]=key.getThree().toString(); } if (!thedate.equals(comparedColumn[0])||!cookie.equals(comparedColumn[1])) { // context.write(new Text(key.getFirst()+" "+key.getSecond().toString()), new Text(key.getThree().toString()+" second"+ " "+comparedColumn[0]+" "+comparedColumn[1]+" "+comparedColumn[2])); houselist.add(key.getThree().toString()); comparedColumn[0]=key.getFirst().toString(); comparedColumn[1]=key.getSecond().toString(); comparedColumn[2]=key.getThree().toString(); } } keyv.set(comparedColumn[0]); //日期 //valuev.set(houselist.toString()); //logger.info(houselist.toString()); //context.write(keyv,valuev); for (int i=0;i<houselist.size()-1;i++) { for (int j=i+1;j<houselist.size();j++) { valuev.set(houselist.get(i)+""+houselist.get(j)); //关联的楼盘 context.write(keyv,valuev); } } } } public static void main(String[] args) throws Exception { Configuration conf = new Configuration(); String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs(); if (otherArgs.length != 2) { System.err.println("Usage: wordcount <in> <out>"); System.exit(2); } FileSystem fstm = FileSystem.get(conf); Path outDir = new Path(otherArgs[1]); fstm.delete(outDir, true); conf.set("mapred.textoutputformat.separator", "\t"); //reduce输出时key value中间的分隔符 Job job = new Job(conf, "HouseMergeAndSplit"); job.setNumReduceTasks(4); job.setJarByClass(HouseMergeAndSplit.class); job.setMapperClass(TokenizerMapper.class); job.setMapOutputKeyClass(TextPair.class); job.setMapOutputValueClass(Text.class); // 设置partition job.setPartitionerClass(Partitioner1.class); // 在分区之后按照指定的条件分组 job.setGroupingComparatorClass(Comp1.class); // 设置reduce // 设置reduce的输出 job.setReducerClass(IntSumReducer.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(Text.class); //job.setNumReduceTasks(18); FileInputFormat.addInputPath(job, new Path(otherArgs[0])); FileOutputFormat.setOutputPath(job, new Path(otherArgs[1])); System.exit(job.waitForCompletion(true) ? 0 : 1); }}TextPair
import java.io.DataInput;import java.io.DataOutput;import java.io.IOException;import org.apache.hadoop.io.Text;import org.apache.hadoop.io.WritableComparable;public class TextPair implements WritableComparable<TextPair> {private Text first;private Text second;private Text three;public TextPair() { set(new Text(), new Text(),new Text());}public TextPair(String first, String second,String three) { set(new Text(first), new Text(second),new Text(three));}public TextPair(Text first, Text second,Text Three) { set(first, second,three);}public void set(Text first, Text second,Text three) { this.first = first; this.second = second; this.three=three;}public Text getFirst() { return first;}public Text getSecond() { return second;}public Text getThree() { return three;}public void write(DataOutput out) throws IOException { first.write(out); second.write(out); three.write(out);}public void readFields(DataInput in) throws IOException { first.readFields(in); second.readFields(in); three.readFields(in);}public int compareTo(TextPair tp) { int cmp = first.compareTo(tp.first); if (cmp != 0) { return cmp; } cmp= second.compareTo(tp.second); if (cmp != 0) { return cmp; } return three.compareTo(tp.three);}}
TextPairSecond
import java.io.DataInput;import java.io.DataOutput;import java.io.IOException;import org.apache.hadoop.io.FloatWritable;import org.apache.hadoop.io.Text;import org.apache.hadoop.io.WritableComparable;public class TextPairSecond implements WritableComparable<TextPairSecond> {private Text first;private FloatWritable second;public TextPairSecond() { set(new Text(), new FloatWritable());}public TextPairSecond(String first, float second) { set(new Text(first), new FloatWritable(second));}public TextPairSecond(Text first, FloatWritable second) { set(first, second);}public void set(Text first, FloatWritable second) { this.first = first; this.second = second;}public Text getFirst() { return first;}public FloatWritable getSecond() { return second;}public void write(DataOutput out) throws IOException { first.write(out); second.write(out);}public void readFields(DataInput in) throws IOException { first.readFields(in); second.readFields(in);}public int compareTo(TextPairSecond tp) { int cmp = first.compareTo(tp.first); if (cmp != 0) { return cmp; } return second.compareTo(tp.second);}}
第二个MR
import java.io.IOException;import java.text.SimpleDateFormat;import java.util.ArrayList;import java.util.Date;import org.apache.hadoop.conf.Configuration;import org.apache.hadoop.fs.FileSystem;import org.apache.hadoop.fs.Path;import org.apache.hadoop.io.IntWritable;import org.apache.hadoop.io.LongWritable;import org.apache.hadoop.io.NullWritable;import org.apache.hadoop.io.Text;import org.apache.hadoop.io.WritableComparable;import org.apache.hadoop.io.WritableComparator;import org.apache.hadoop.mapred.OutputCollector;import org.apache.hadoop.mapreduce.Job;import org.apache.hadoop.mapreduce.Mapper;import org.apache.hadoop.mapreduce.Partitioner;import org.apache.hadoop.mapreduce.Reducer;import org.apache.hadoop.mapreduce.Mapper.Context;import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;import org.apache.hadoop.util.GenericOptionsParser;import org.apache.log4j.Logger;/* * 统计楼盘之间共同出现的次数 * 输入: * 日期 楼盘1 楼盘2 * * 输出： * 日期 楼盘1 楼盘2 共同出现的次数 * */public class HouseCount { public static class TokenizerMapper extends Mapper<LongWritable, Text, Text, IntWritable>{ IntWritable iw=new IntWritable(1); public void map(LongWritable key, Text value, Context context ) throws IOException, InterruptedException { context.write(value, iw); } } public static class IntSumReducer extends Reducer<Text,IntWritable,Text,IntWritable> { IntWritable result=new IntWritable(); public void reduce(Text key, Iterable<IntWritable> values, Context context ) throws IOException, InterruptedException { int sum=0; for (IntWritable iw:values) { sum+=iw.get(); } result.set(sum); context.write(key, result); } } public static void main(String[] args) throws Exception { Configuration conf = new Configuration(); String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs(); if (otherArgs.length != 2) { System.err.println("Usage: wordcount <in> <out>"); System.exit(2); } FileSystem fstm = FileSystem.get(conf); Path outDir = new Path(otherArgs[1]); fstm.delete(outDir, true); conf.set("mapred.textoutputformat.separator", "\t"); //reduce输出时key value中间的分隔符 Job job = new Job(conf, "HouseCount"); job.setNumReduceTasks(2); job.setJarByClass(HouseCount.class); job.setMapperClass(TokenizerMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(IntWritable.class); // 设置reduce // 设置reduce的输出 job.setReducerClass(IntSumReducer.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(IntWritable.class); //job.setNumReduceTasks(18); FileInputFormat.addInputPath(job, new Path(otherArgs[0])); FileOutputFormat.setOutputPath(job, new Path(otherArgs[1])); System.exit(job.waitForCompletion(true) ? 0 : 1); }}
第三个MR
import java.io.IOException;import java.text.ParseException;import java.text.SimpleDateFormat;import java.util.ArrayList;import java.util.Calendar;import java.util.Date;import org.apache.hadoop.conf.Configuration;import org.apache.hadoop.fs.FileSystem;import org.apache.hadoop.fs.Path;import org.apache.hadoop.io.FloatWritable;import org.apache.hadoop.io.IntWritable;import org.apache.hadoop.io.LongWritable;import org.apache.hadoop.io.NullWritable;import org.apache.hadoop.io.Text;import org.apache.hadoop.io.WritableComparable;import org.apache.hadoop.io.WritableComparator;import org.apache.hadoop.mapred.OutputCollector;import org.apache.hadoop.mapreduce.Job;import org.apache.hadoop.mapreduce.Mapper;import org.apache.hadoop.mapreduce.Partitioner;import org.apache.hadoop.mapreduce.Reducer;import org.apache.hadoop.mapreduce.Mapper.Context;import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;import org.apache.hadoop.util.GenericOptionsParser;import org.apache.log4j.Logger;/* * 汇总近三个月统计楼盘之间共同出现的次数,考虑衰减系数， 并最后a b 转成 b a输出一次 * 输入: * 日期 楼盘1 楼盘2 共同出现的次数 * * 输出 * 楼盘1 楼盘2 共同出现的次数（考虑了衰减系数，每天的衰减系数不一样) * */public class HouseCountHz { public static class HouseCountHzMapper extends Mapper<LongWritable, Text, Text, FloatWritable>{ Text keyv=new Text();FloatWritable valuev=new FloatWritable(); public void map(LongWritable key, Text value, Context context ) throws IOException, InterruptedException { String[] s=value.toString().split("\t"); keyv.set(s[1]+""+s[2]);//楼盘1,楼盘2 Calendar date1=Calendar.getInstance(); Calendar d2=Calendar.getInstance(); Date b = null; SimpleDateFormat sdf=new SimpleDateFormat("yyyy-MM-dd"); try { b=sdf.parse(s[0]); } catch (ParseException e) { e.printStackTrace(); } d2.setTime(b); long n=date1.getTimeInMillis(); long birth=d2.getTimeInMillis(); long sss=n-birth; int day=(int)((sss)/(3600*24*1000)); //该条记录的日期与当前日期的日期差 float factor=1/(1+(float)(day-1)/10); //衰减系数 valuev.set(Float.parseFloat(s[3])*factor); context.write(keyv, valuev); } } public static class HouseCountHzReducer extends Reducer<Text,FloatWritable,Text,FloatWritable> { FloatWritable result=new FloatWritable(); Text keyreverse=new Text(); public void reduce(Text key, Iterable<FloatWritable> values, Context context ) throws IOException, InterruptedException { float sum=0; for (FloatWritable iw:values) { sum+=iw.get(); } result.set(sum); String[] keys=key.toString().split("\t"); keyreverse.set(keys[1]+""+keys[0]); context.write(key, result); context.write(keyreverse, result); } } public static void main(String[] args) throws Exception { Configuration conf = new Configuration(); String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs(); if (otherArgs.length != 2) { System.err.println("Usage: wordcount <in> <out>"); System.exit(2); } FileSystem fstm = FileSystem.get(conf); Path outDir = new Path(otherArgs[1]); fstm.delete(outDir, true); conf.set("mapred.textoutputformat.separator", "\t"); //reduce输出时key value中间的分隔符 Job job = new Job(conf, "HouseCountHz"); job.setNumReduceTasks(2); job.setJarByClass(HouseCountHz.class); job.setMapperClass(HouseCountHzMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(FloatWritable.class); // 设置reduce // 设置reduce的输出 job.setReducerClass(HouseCountHzReducer.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(FloatWritable.class); //job.setNumReduceTasks(18); FileInputFormat.addInputPath(job, new Path(otherArgs[0])); FileOutputFormat.setOutputPath(job, new Path(otherArgs[1])); System.exit(job.waitForCompletion(true) ? 0 : 1); }}
第四个MR
import java.io.IOException;import java.util.Iterator;import org.apache.hadoop.conf.Configuration;import org.apache.hadoop.fs.FileSystem;import org.apache.hadoop.fs.Path;import org.apache.hadoop.io.FloatWritable;import org.apache.hadoop.io.LongWritable;import org.apache.hadoop.io.Text;import org.apache.hadoop.io.WritableComparable;import org.apache.hadoop.io.WritableComparator;import org.apache.hadoop.mapreduce.Job;import org.apache.hadoop.mapreduce.Mapper;import org.apache.hadoop.mapreduce.Partitioner;import org.apache.hadoop.mapreduce.Reducer;import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;import org.apache.hadoop.util.GenericOptionsParser;/* * 输入数据: * 楼盘1 楼盘2 共同出现的次数 * * 输出数据 * 楼盘1 楼盘2,楼盘3,楼盘4 (按次数排序) */public class HouseRowToCol {public static class Partitioner1 extends Partitioner<TextPairSecond, Text> { @Override //分区 public int getPartition(TextPairSecond key, Text value, int numParititon) { return Math.abs((new Text(key.getFirst().toString()+key.getSecond().toString())).hashCode() * 127) % numParititon; }}//分组 public static class Comp1 extends WritableComparator { public Comp1() { super(TextPairSecond.class, true); } @SuppressWarnings("unchecked") public int compare(WritableComparable a, WritableComparable b) { TextPairSecond t1 = (TextPairSecond) a; TextPairSecond t2 = (TextPairSecond) b; return t1.getFirst().compareTo(t2.getFirst()); }} //排序 public static class KeyComp extends WritableComparator { public KeyComp() { super(TextPairSecond.class, true); } @SuppressWarnings("unchecked") public int compare(WritableComparable a, WritableComparable b) { TextPairSecond t1 = (TextPairSecond) a; TextPairSecond t2 = (TextPairSecond) b; int comp= t1.getFirst().compareTo(t2.getFirst()); if (comp!=0) return comp; return -t1.getSecond().compareTo(t2.getSecond()); }} public static class HouseRowToColMapper extends Mapper<LongWritable, Text, TextPairSecond, Text>{ Text houseid1=new Text(); Text houseid2=new Text(); FloatWritable weight=new FloatWritable(); public void map(LongWritable key, Text value, Context context ) throws IOException, InterruptedException { String s[]=value.toString().split("\t"); weight.set(Float.parseFloat(s[2])); houseid1.set(s[0]); houseid2.set(s[1]); TextPairSecond tp=new TextPairSecond(houseid1,weight); context.write(tp, houseid2); } } public static class HouseRowToColReducer extends Reducer<TextPairSecond,Text,Text,Text> { Text valuev=new Text(); public void reduce(TextPairSecond key, Iterable<Text> values, Context context ) throws IOException, InterruptedException { Text keyv=key.getFirst(); Iterator<Text> it=values.iterator(); StringBuilder sb=new StringBuilder(it.next().toString()); while(it.hasNext()) { sb.append(","+it.next().toString()); } valuev.set(sb.toString()); context.write(keyv, valuev); } } public static void main(String[] args) throws Exception { Configuration conf = new Configuration(); String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs(); if (otherArgs.length != 2) { System.err.println("Usage: wordcount <in> <out>"); System.exit(2); } FileSystem fstm = FileSystem.get(conf); Path outDir = new Path(otherArgs[1]); fstm.delete(outDir, true); conf.set("mapred.textoutputformat.separator", "\t"); //reduce输出时key value中间的分隔符 Job job = new Job(conf, "HouseRowToCol"); job.setNumReduceTasks(4); job.setJarByClass(HouseRowToCol.class); job.setMapperClass(HouseRowToColMapper.class); job.setMapOutputKeyClass(TextPairSecond.class); job.setMapOutputValueClass(Text.class); // 设置partition job.setPartitionerClass(Partitioner1.class); // 在分区之后按照指定的条件分组 job.setGroupingComparatorClass(Comp1.class); job.setSortComparatorClass(KeyComp.class); // 设置reduce // 设置reduce的输出 job.setReducerClass(HouseRowToColReducer.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(Text.class); //job.setNumReduceTasks(18); FileInputFormat.addInputPath(job, new Path(otherArgs[0])); FileOutputFormat.setOutputPath(job, new Path(otherArgs[1])); System.exit(job.waitForCompletion(true) ? 0 : 1); }}