package cn.test.business;import java.io.File;import java.sql.Connection;import java.sql.ResultSet;import java.sql.SQLException;import java.sql.Statement;import java.util.Calendar;import java.util.HashSet;import java.util.Iterator;import java.util.LinkedHashMap;import java.util.List;import java.util.TreeMap;import org.apache.commons.logging.Log;import org.apache.commons.logging.LogFactory;import org.apache.mahout.cf.taste.eval.RecommenderBuilder;import org.apache.mahout.cf.taste.impl.common.LongPrimitiveIterator;import org.apache.mahout.cf.taste.model.DataModel;import org.apache.mahout.cf.taste.recommender.RecommendedItem;import redis.clients.jedis.Jedis;import redis.clients.jedis.JedisPool; public class OldUserBus { static String outMonth = Global.getConfig("outMonth"); static JedisPool pool = JedisUtil.getPool(); static Jedis jedis = pool.getResource(); static float itemScore = 0.03f; // 如果算法没有推荐分数，需要计算的分数 static float priceLat = 0.001f; // 计算的价格 static float priceMid = 259; // 地区房价价格临界点 static float dayLat = 0.001f; // 天数越大，越不推荐 static float smallAreaCodeScore = 0.011f; // 该地区值多少分，地区基本分数 static float roomLat = 0.9f; // 非最佳房间数相 static float sitLat = 0.9f; // 非最佳厅数 static int roomNum = 3; // 最佳推荐房间数 static int sitNum = 1; // 最佳推荐厅间数 /* *teseJuint对算法结果提取出来，进行二次运算，添加各种纬度进行结果运算 * * itemLoglikelihood */ public static int adLatite(long exitId, LinkedHashMap<Integer, Float> tmpMap, int smallAreaCode) { String udid = UserBus.catRedis2udid(Integer.valueOf(exitId + "")); int Id = 0; double price = 0.0; String roomType = ""; String createTime = ""; long dayNum = 0; double score = 0.0; String date = DateUtils.getDate("yyyy/MM/dd HH:mm:s"); try { // 根据uid，找到用户之前没有看到房源，进行一个排分,再按照一个时间排序 String sql = "SELECT DISTINCT h.`Id`, h.`Price` ,h.`roomType` ,h.`createTime` FROM hw_h_click_log h " + "WHERE h.`smallAreaCode`=" + smallAreaCode + " AND h.`udid` !='" + udid + "' ORDER BY h.`createTime` DESC ,h.`Price` LIMIT 10 "; ResultSet res = SqlUtils.result(sql); TreeMap<Double, Double> tmp = new TreeMap<Double, Double>(); while (res.next()) { Id = res.getInt(1); price = res.getDouble(2); roomType = res.getString(3); createTime = res.getDate(4) + ""; String[] spilt = roomType.replace("厅", "").split("室"); // System.out.println(spilt[0] +"--"+spilt[1]); int roomInt = Integer.valueOf(spilt[0]); int sitInt = Integer.valueOf(spilt[1]); Float itemLat = tmpMap.get(res.getInt(1)); if (null == itemLat || StringUtils.isBlank(itemLat + "")) { itemLat = itemScore; } // System.out.println("-result ::" +java.lang.Math.abs(result)); dayNum = DateUtils.pastDays(DateUtils.parseDate(createTime));// 计算多少天数 double priceScore = price - priceMid; double dayScore = dayNum * dayLat; // 发布天数乘以 天的纬度数 score = itemLat + smallAreaCodeScore - (priceScore * priceLat) - dayScore;// 得到推荐分数 System.out.println("--itemLat：：：" + itemLat + "--dayScore：：：" + dayNum * dayLat + "--score:::" + score + "--priceScore:::" + priceScore * priceLat); tmp.put(score, (double) Id); } Iterator<Double> iterator_2 = tmp.keySet().iterator(); int sort = 0; /* * DateUtils.getDate("yyyy/MM/dd HH:mm:s") */ while (iterator_2.hasNext()) { sort++; Object key = iterator_2.next(); System.out.println(sort + "::tmp.get(key) is :" + tmp.get(key) + "--- " + key); sql = "INSERT INTO `hw_h_old_recom` (`udid`,`Id`,`smallAreaCode`,`sort`,`score`, `createTime` ) VALUES " + "('" + udid + "'," + tmp.get(key) + "," + smallAreaCode + " , " + sort + " , '" + key + "' , '" + DateUtils.getDate("yyyy/MM/dd HH:mm:s") + "')"; // boolean bb=SqlUtils.insert(sql); // System.out.println("--insert---" +bb); } } catch (Exception e) { // TODO Auto-generated catch block e.printStackTrace(); } return 0; } /* *teseJuint调用本地算法进行计算 * * itemLoglikelihood */ public static int userCityLog(int smallAreaCode) { try { String pvCsv = "datafile//" + smallAreaCode + "pv.csv"; String Csv = "datafile//" + smallAreaCode + ".csv"; DataModel dataModel = RecommendFactory.buildDataModelNoPref(pvCsv); RecommenderBuilder userCity = RecommenderEvaluator .userCityBlock(dataModel); LongPrimitiveIterator iter = dataModel.getUserIDs(); while (iter.hasNext()) { long uid = iter.nextLong(); System.out.print("userCityBlock =>"); RecommenderFilterOutdateResult.filterOutdate(uid, userCity, dataModel, Csv); List<RecommendedItem> recommendations = RecommenderFilterOutdateResult .filterOutdate(uid, userCity, dataModel, Csv); boolean skip = false; if (!skip || recommendations.size() > 0) { System.out.printf("uid:%s,", uid); for (RecommendedItem recommendation : recommendations) { System.out.printf("(%s,%f)", recommendation.getItemID(), recommendation.getValue()); } System.out.println(); } } } catch (Exception e) { e.printStackTrace(); } return 0; } /* *teseJuint调用本地算法进行计算 * * itemLoglikelihood */ public static int useitemLog(int smallAreaCode) { LinkedHashMap<Integer, LinkedHashMap<Integer, Float>> map = new LinkedHashMap<Integer, LinkedHashMap<Integer, Float>>(); try { areaImport(smallAreaCode); String pvCsv = "datafile//" + smallAreaCode + "pv.csv"; String Csv = "datafile//" + smallAreaCode + ".csv"; DataModel dataModel = RecommendFactory.buildDataModelNoPref(pvCsv); RecommenderBuilder itemLog = RecommenderEvaluator.itemLoglikelihood(dataModel); LongPrimitiveIterator iter = dataModel.getUserIDs(); while (iter.hasNext()) { long uid = iter.nextLong(); System.out.print("itemLoglikelihood=>"); List<RecommendedItem> recommendations = RecommenderFilterOutdateResult .filterOutdate(uid, itemLog, dataModel, Csv); boolean skip = false; if (!skip || recommendations.size() > 0) { System.out.printf("uid:%s,", uid); LinkedHashMap<Integer, Float> tmpMap = new LinkedHashMap<Integer, Float>(); for (RecommendedItem recommendation : recommendations) { System.out.printf("(%s,%f)",recommendation.getItemID(), recommendation.getValue()); tmpMap.put((int) recommendation.getItemID(),recommendation.getValue()); } map.put((int) uid, tmpMap); adLatite(uid, tmpMap, smallAreaCode);// 添加纬度进行运算 System.out.println(); } } } catch (Exception e) { e.printStackTrace(); } return 0; } /* *teseJuint根据itemLoglikelihood算法需要，从 我们的mysql日志表里再导出csv格式的数据 * * 这两个csv文件，做算法匹配 pvCsv 用户id+，+id Csv 房源id+房源时间+房源价格 * * 为了减轻压力，都是分地区而导，分地区而计算PV总数 */ public static int areaImport(int smallAreaCode) { try { String sql = "SELECT h.`udid`, h.`Id` FROM hw_h_click_log h WHERE h.`smallAreaCode`="+ smallAreaCode; ResultSet res = SqlUtils.result(sql); String result = ""; String Ids = ""; HashSet<Integer> set = new HashSet<Integer>(); while (res.next()) { // System.out.println(res.getString(1) +"," +res.getInt(2)); int changeId = UserBus.catUdidInsert(res.getString(1)); result += changeId + "," + res.getInt(2) + "\n"; set.add(res.getInt(2)); } boolean bb = FileUtil.writeTxtFile(result, new File("datafile//" + smallAreaCode + "pv.csv")); result = ""; Iterator<Integer> iterator = set.iterator(); while (iterator.hasNext()) { Ids += iterator.next() + ","; } sql = "SELECT DISTINCT h.`Id` ,h.`createTime` ,h.`Price` FROM hw_h_click_log h WHERE h.`Id` IN (" + Ids.substring(0, Ids.length() - 1) + ")"; res = SqlUtils.result(sql); while (res.next()) { // System.out.println(res.getString(1) +","+res.getDate(2)+"," // +res.getDouble(3)); result += res.getString(1) + "," + res.getDate(2) + "," + res.getDouble(3) + "\n"; } boolean bb = FileUtil.writeTxtFile(result, new File( "datafile//" + smallAreaCode + ".csv")); System.out.println(bb + "----" + bb); } catch (Exception e) { e.printStackTrace(); } return 0; } /* *teseJuint删除所有过期的推荐房源 */ public static int dropArea(String delDate) { try { String sql = "DELETE FROM hw_h_new_recom WHERE hw_h_new_recom.`createTime` < '" + delDate + "'"; ResultSet res = SqlUtils.result(sql); } catch (Exception e) { // TODO Auto-generated catch block e.printStackTrace(); return SysStatus.TTE_FAILED; } return SysStatus.TTE_SUCCESSED; } /* *teseJuint遍历所有的地区 ,特别针对老用户的 */ public static int sortAreaOld() { int smallAreaCode = 0; try { String sql = "SELECT DISTINCT h.`smallAreaCode` FROM hw_h_click_log h"; ResultSet res = SqlUtils.result(sql); while (res.next()) { smallAreaCode = res.getInt(1); if (0 != res.getInt(1)&& StringUtils.isNotBlank(smallAreaCode + "")) { useitemLog(smallAreaCode); } } } catch (Exception e) { // TODO Auto-generated catch block e.printStackTrace(); return SysStatus.TTE_FAILED; } return SysStatus.TTE_SUCCESSED; } /* * 入查：地区id 出查：地区的平均房价 * * 这个地方有待改进 */ public static double areaAvgPrice(int smallAreaCode) { Connection conn = null; Statement stmt = null; ResultSet res = null; double price = 0.0; int i = 0; try { conn = MqjdbcUtils.getConnection(); stmt = conn.createStatement(); res = stmt.executeQuery("SELECT DISTINCT h.`Id` ,h.`Price` FROM hw_h_click_log h WHERE h.`smallAreaCode` =" + smallAreaCode); while (res.next()) { // System.out.println(res.getInt(1) +"--" +res.getDouble(2)); price += res.getDouble(2); i++; } System.out.println(price / i + "执行运行结�?::" + smallAreaCode); return price / i; } catch (SQLException e) { // TODO Auto-generated catch block e.printStackTrace(); return price / i; } } public static void main(String[] args) { sortAreaOld(); // areaImport(11891); // useitemLog(11891); // userCityLog(11891); // adLatite(exitId, tmpMap, smallAreaCode); }
4103
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
