排序新公式

recommend-server-service/src/main/java/com/tzld/piaoquan/recommend/server/common/base/RankItem.java

@@ -13,6 +13,8 @@ public class RankItem implements Comparable<RankItem> {
 
     // featureMap中保存所有的特征
     public Map<String, String> featureMap = new HashMap<>();
+    public Map<String, String> itemBasicFeature = new HashMap<>();
+    public Map<String, Map<String, Double>> itemRealTimeFeature = new HashMap<>();
     public long videoId;
     private double score; // 记录最终的score
     private Video video;

recommend-server-service/src/main/java/com/tzld/piaoquan/recommend/server/service/rank/strategy/RankStrategy4RegionMergeModel.java

@@ -0,0 +1,372 @@
+package com.tzld.piaoquan.recommend.server.service.rank.strategy;
+
+
+import com.alibaba.fastjson.JSONObject;
+import com.ctrip.framework.apollo.spring.annotation.ApolloJsonValue;
+import com.google.common.reflect.TypeToken;
+import com.tzld.piaoquan.recommend.server.common.base.RankItem;
+import com.tzld.piaoquan.recommend.server.model.Video;
+import com.tzld.piaoquan.recommend.server.service.rank.RankParam;
+import com.tzld.piaoquan.recommend.server.service.rank.RankService;
+import com.tzld.piaoquan.recommend.server.service.rank.extractor.RankExtractorItemFeature;
+import com.tzld.piaoquan.recommend.server.service.rank.extractor.RankExtractorUserFeature;
+import com.tzld.piaoquan.recommend.server.service.recall.strategy.*;
+import com.tzld.piaoquan.recommend.server.service.score.ScorerUtils;
+import com.tzld.piaoquan.recommend.server.util.CommonCollectionUtils;
+import com.tzld.piaoquan.recommend.server.util.JSONUtils;
+import lombok.extern.slf4j.Slf4j;
+import org.apache.commons.collections4.CollectionUtils;
+import org.apache.commons.lang3.math.NumberUtils;
+import org.springframework.data.redis.connection.RedisConnectionFactory;
+import org.springframework.data.redis.connection.RedisStandaloneConfiguration;
+import org.springframework.data.redis.connection.jedis.JedisConnectionFactory;
+import org.springframework.data.redis.core.RedisTemplate;
+import org.springframework.data.redis.serializer.StringRedisSerializer;
+import org.springframework.stereotype.Service;
+
+import java.text.SimpleDateFormat;
+import java.util.*;
+import java.util.stream.Collectors;
+
+/**
+ * @author zhangbo
+ * @desc 地域召回融合
+ */
+@Service
+@Slf4j
+public class RankStrategy4RegionMergeModel extends RankService {
+    @ApolloJsonValue("${video.model.weightv3:}")
+    private Map<String, Double> mergeWeight;
+    final private String CLASS_NAME = this.getClass().getSimpleName();
+    public void duplicate(Set<Long> setVideo, List<Video> videos){
+        Iterator<Video> iterator = videos.iterator();
+        while(iterator.hasNext()){
+            Video v = iterator.next();
+            if (setVideo.contains(v.getVideoId())){
+                iterator.remove();
+            }else{
+                setVideo.add(v.getVideoId());
+            }
+        }
+    }
+    @Override
+    public List<Video> mergeAndRankRovRecall(RankParam param) {
+        //-------------------地域内部融合+去重复-------------------
+        List<Video> rovRecallRank = new ArrayList<>();
+        List<Video> v1 = extractAndSort(param, RegionRealtimeRecallStrategyV1.PUSH_FORM);
+        List<Video> v2 = extractAndSort(param, RegionRealtimeRecallStrategyV2.PUSH_FORM);
+        List<Video> v3 = extractAndSort(param, RegionRealtimeRecallStrategyV3.PUSH_FORM);
+        List<Video> v4 = extractAndSort(param, RegionRealtimeRecallStrategyV4.PUSH_FORM);
+        Set<Long> setVideo = new HashSet<>();
+        this.duplicate(setVideo, v1);
+        this.duplicate(setVideo, v2);
+        this.duplicate(setVideo, v3);
+        this.duplicate(setVideo, v4);
+        //-------------------地域 sim returnv2 融合+去重复-------------------
+        List<Video> v5 = extractAndSort(param, SimHotVideoRecallStrategy.PUSH_FORM);
+        List<Video> v6 = extractAndSort(param, ReturnVideoRecallStrategy.PUSH_FORM);
+        this.duplicate(setVideo, v5);
+        this.duplicate(setVideo, v6);
+
+//        rovRecallRank.addAll(v1);
+//        rovRecallRank.addAll(v2);
+//        rovRecallRank.addAll(v3);
+//        rovRecallRank.addAll(v4);
+//        rovRecallRank.addAll(v5);
+//        rovRecallRank.addAll(v6);
+
+        rovRecallRank.addAll(v1.subList(0, Math.min(20, v1.size())));
+        rovRecallRank.addAll(v2.subList(0, Math.min(15, v2.size())));
+        rovRecallRank.addAll(v3.subList(0, Math.min(10, v3.size())));
+        rovRecallRank.addAll(v4.subList(0, Math.min(5, v4.size())));
+        rovRecallRank.addAll(v5.subList(0, Math.min(10, v5.size())));
+        rovRecallRank.addAll(v6.subList(0, Math.min(10, v6.size())));
+
+        //-------------------排-------------------
+        //-------------------序-------------------
+        //-------------------逻-------------------
+        //-------------------辑-------------------
+//        List<String> videoIdKeys = rovRecallRank.stream()
+//                .map(t -> param.getRankKeyPrefix() + t.getVideoId())
+//                .collect(Collectors.toList());
+//        List<String> videoScores = this.redisTemplate.opsForValue().multiGet(videoIdKeys);
+//        log.info("rank mergeAndRankRovRecall videoIdKeys={}, videoScores={}", JSONUtils.toJson(videoIdKeys),
+//                JSONUtils.toJson(videoScores));
+//        if (CollectionUtils.isNotEmpty(videoScores)
+//                && videoScores.size() == rovRecallRank.size()) {
+//            for (int i = 0; i < videoScores.size(); i++) {
+//                rovRecallRank.get(i).setSortScore(NumberUtils.toDouble(videoScores.get(i), 0.0));
+//            }
+//            Collections.sort(rovRecallRank, Comparator.comparingDouble(o -> -o.getSortScore()));
+//        }
+
+        // 1 模型分 todo
+        List<String> rtFeaPart = new ArrayList<>();
+        List<RankItem> videos = model(rovRecallRank, param, rtFeaPart);
+        Calendar calendar = Calendar.getInstance();
+        String date = new SimpleDateFormat("yyyyMMdd").format(calendar.getTime());
+        String hour = new SimpleDateFormat("HH").format(calendar.getTime());
+        String rtFeaPart1h = date + hour;
+        if (rtFeaPart != null){
+            if (rtFeaPart.get(1) != null){
+                rtFeaPart1h = rtFeaPart.get(1);
+            }
+        }
+        // 2 统计分 todo
+        for (RankItem video : videos){
+            Map<String, String> itemBasicMap = video.getItemBasicFeature();
+            Map<String, Map<String, Double>> itemRealMap = video.getItemRealTimeFeature();
+
+        }
+        // 3 融合公式 todo
+
+
+        return videos;
+    }
+
+    public List<RankItem> model(List<Video> videos, RankParam param,
+                             List<String> rtFeaPart){
+        List<RankItem> result = new ArrayList<>();
+        if (videos.isEmpty()){
+            return result;
+        }
+
+        RedisStandaloneConfiguration redisSC = new RedisStandaloneConfiguration();
+        redisSC.setPort(6379);
+        redisSC.setPassword("Wqsd@2019");
+        redisSC.setHostName("r-bp1pi8wyv6lzvgjy5z.redis.rds.aliyuncs.com");
+        RedisConnectionFactory connectionFactory = new JedisConnectionFactory(redisSC);
+        RedisTemplate<String, String> redisTemplate = new RedisTemplate<>();
+        redisTemplate.setConnectionFactory(connectionFactory);
+        redisTemplate.setDefaultSerializer(new StringRedisSerializer());
+        redisTemplate.afterPropertiesSet();
+
+        // 0: 场景特征处理
+        Map<String, String> sceneFeatureMap =  this.getSceneFeature(param);
+
+        // 1: user特征处理
+        Map<String, String> userFeatureMap = new HashMap<>();
+        if (param.getMid() != null && !param.getMid().isEmpty()){
+            String midKey = "user_info_4video_" + param.getMid();
+            String userFeatureStr = redisTemplate.opsForValue().get(midKey);
+            if (userFeatureStr != null){
+                try{
+                    userFeatureMap = JSONUtils.fromJson(userFeatureStr,
+                            new TypeToken<Map<String, String>>() {},
+                            userFeatureMap);
+                }catch (Exception e){
+                    log.error(String.format("parse user json is wrong in {} with {}", this.CLASS_NAME, e));
+                }
+            }else{
+                JSONObject obj = new JSONObject();
+                obj.put("name", "user_key_in_model_is_null");
+                log.info(obj.toString());
+//                return videos;
+            }
+        }
+        final Set<String> userFeatureSet = new HashSet<>(Arrays.asList(
+                "machineinfo_brand", "machineinfo_model", "machineinfo_platform", "machineinfo_system",
+                "u_1day_exp_cnt", "u_1day_click_cnt", "u_1day_share_cnt", "u_1day_return_cnt",
+                "u_3day_exp_cnt", "u_3day_click_cnt", "u_3day_share_cnt", "u_3day_return_cnt"
+        ));
+        Iterator<Map.Entry<String, String>> iterator = userFeatureMap.entrySet().iterator();
+        while (iterator.hasNext()) {
+            Map.Entry<String, String> entry = iterator.next();
+            if (!userFeatureSet.contains(entry.getKey())) {
+                iterator.remove();
+            }
+        }
+        Map<String, String> f1 = RankExtractorUserFeature.getOriginFeature(userFeatureMap,
+                new HashSet<String>(Arrays.asList(
+                        "machineinfo_brand", "machineinfo_model", "machineinfo_platform", "machineinfo_system"
+                ))
+        );
+        Map<String, String> f2 = RankExtractorUserFeature.getUserRateFeature(userFeatureMap);
+        Map<String, String> f3 = RankExtractorUserFeature.cntFeatureChange(userFeatureMap,
+                new HashSet<String>(Arrays.asList(
+                        "u_1day_exp_cnt", "u_1day_click_cnt", "u_1day_share_cnt", "u_1day_return_cnt",
+                        "u_3day_exp_cnt", "u_3day_click_cnt", "u_3day_share_cnt", "u_3day_return_cnt"
+                ))
+        );
+        f1.putAll(f2);
+        f1.putAll(f3);
+        log.info("userFeature in model = {}", JSONUtils.toJson(f1));
+
+        // 2-1: item特征处理
+        final Set<String> itemFeatureSet = new HashSet<>(Arrays.asList(
+                "total_time", "play_count_total",
+                "i_1day_exp_cnt", "i_1day_click_cnt", "i_1day_share_cnt", "i_1day_return_cnt",
+                "i_3day_exp_cnt", "i_3day_click_cnt", "i_3day_share_cnt", "i_3day_return_cnt"
+        ));
+
+        List<RankItem> rankItems = CommonCollectionUtils.toList(videos, RankItem::new);
+        List<Long> videoIds = CommonCollectionUtils.toListDistinct(videos, Video::getVideoId);
+        List<String> videoFeatureKeys = videoIds.stream().map(r-> "video_info_" + r)
+                .collect(Collectors.toList());
+        List<String> videoFeatures = redisTemplate.opsForValue().multiGet(videoFeatureKeys);
+        if (videoFeatures != null){
+            for (int i=0; i<videoFeatures.size(); ++i){
+                String vF = videoFeatures.get(i);
+                Map<String, String> vfMap = new HashMap<>();
+                if (vF == null){
+                    continue;
+                }
+                try{
+                    vfMap = JSONUtils.fromJson(vF, new TypeToken<Map<String, String>>() {}, vfMap);
+                    rankItems.get(i).setItemBasicFeature(vfMap);
+                    Iterator<Map.Entry<String, String>> iteratorIn = vfMap.entrySet().iterator();
+                    while (iteratorIn.hasNext()) {
+                        Map.Entry<String, String> entry = iteratorIn.next();
+                        if (!itemFeatureSet.contains(entry.getKey())) {
+                            iteratorIn.remove();
+                        }
+                    }
+                    Map<String, String> f4 = RankExtractorItemFeature.getItemRateFeature(vfMap);
+                    Map<String, String> f5 = RankExtractorItemFeature.cntFeatureChange(vfMap,
+                            new HashSet<String>(Arrays.asList(
+                                    "total_time", "play_count_total",
+                                    "i_1day_exp_cnt", "i_1day_click_cnt", "i_1day_share_cnt", "i_1day_return_cnt",
+                                    "i_3day_exp_cnt", "i_3day_click_cnt", "i_3day_share_cnt", "i_3day_return_cnt"))
+                    );
+                    f4.putAll(f5);
+                    rankItems.get(i).setFeatureMap(f4);
+                }catch (Exception e){
+                    log.error(String.format("parse video json is wrong in {} with {}", this.CLASS_NAME, e));
+                }
+            }
+        }
+        // 2-2: item 实时特征处理
+        List<String> rtFeaPartKey = new ArrayList<>(Arrays.asList("item_rt_fea_1day_partition", "item_rt_fea_1h_partition"));
+        List<String> rtFeaPartKeyResult = this.redisTemplate.opsForValue().multiGet(rtFeaPartKey);
+        rtFeaPart = rtFeaPartKeyResult;
+        Calendar calendar = Calendar.getInstance();
+        String date = new SimpleDateFormat("yyyyMMdd").format(calendar.getTime());
+        String hour = new SimpleDateFormat("HH").format(calendar.getTime());
+        String rtFeaPart1day = date + hour;
+        String rtFeaPart1h = date + hour;
+        if (rtFeaPartKeyResult != null){
+            if (rtFeaPartKeyResult.get(0) != null){
+                rtFeaPart1day = rtFeaPartKeyResult.get(0);
+            }
+            if (rtFeaPartKeyResult.get(1) != null){
+                rtFeaPart1h = rtFeaPartKeyResult.get(1);
+            }
+        }
+
+        List<String> videoRtKeys1 = videoIds.stream().map(r-> "item_rt_fea_1day_" + r)
+                .collect(Collectors.toList());
+        List<String> videoRtKeys2 = videoIds.stream().map(r-> "item_rt_fea_1h_" + r)
+                .collect(Collectors.toList());
+        videoRtKeys1.addAll(videoRtKeys2);
+        List<String> videoRtFeatures = this.redisTemplate.opsForValue().multiGet(videoRtKeys1);
+
+
+        if (videoRtFeatures != null){
+            int j = 0;
+            for (RankItem item: rankItems){
+                String vF = videoRtFeatures.get(j);
+                ++j;
+                if (vF == null){
+                    continue;
+                }
+                Map<String, String> vfMap = new HashMap<>();
+                Map<String, Map<String, Double>> vfMapNew = new HashMap<>();
+                try{
+                    vfMap = JSONUtils.fromJson(vF, new TypeToken<Map<String, String>>() {}, vfMap);
+                    for (Map.Entry<String, String> entry : vfMap.entrySet()){
+                        String value = entry.getValue();
+                        if (value == null){
+                            continue;
+                        }
+                        String [] var1 = value.split(",");
+                        Map<String, Double> tmp = new HashMap<>();
+                        for (String var2 : var1){
+                            String [] var3 = var2.split(":");
+                            tmp.put(var3[0], Double.valueOf(var3[1]));
+                        }
+                        vfMapNew.put(entry.getKey(), tmp);
+                    }
+                }catch (Exception e){
+                    log.error(String.format("parse video item_rt_fea_1day_ json is wrong in {} with {}", this.CLASS_NAME, e));
+                }
+                Map<String, String> f8 = RankExtractorItemFeature.getItemRealtimeRate(vfMapNew, rtFeaPart1day);
+                item.getFeatureMap().putAll(f8);
+            }
+            for (RankItem item: rankItems){
+                String vF = videoRtFeatures.get(j);
+                ++j;
+                if (vF == null){
+                    continue;
+                }
+                Map<String, String> vfMap = new HashMap<>();
+                Map<String, Map<String, Double>> vfMapNew = new HashMap<>();
+                try{
+                    vfMap = JSONUtils.fromJson(vF, new TypeToken<Map<String, String>>() {}, vfMap);
+
+                    for (Map.Entry<String, String> entry : vfMap.entrySet()){
+                        String value = entry.getValue();
+                        if (value == null){
+                            continue;
+                        }
+                        String [] var1 = value.split(",");
+                        Map<String, Double> tmp = new HashMap<>();
+                        for (String var2 : var1){
+                            String [] var3 = var2.split(":");
+                            tmp.put(var3[0], Double.valueOf(var3[1]));
+                        }
+                        vfMapNew.put(entry.getKey(), tmp);
+                    }
+                    item.setItemRealTimeFeature(vfMapNew);
+                }catch (Exception e){
+                    log.error(String.format("parse video item_rt_fea_1h_ json is wrong in {} with {}", this.CLASS_NAME, e));
+                }
+                Map<String, String> f8 = RankExtractorItemFeature.getItemRealtimeRate(vfMapNew, rtFeaPart1h);
+                item.getFeatureMap().putAll(f8);
+            }
+        }
+
+
+        log.info("ItemFeature = {}", JSONUtils.toJson(videoFeatures));
+
+
+
+        List<RankItem> rovRecallScore = ScorerUtils.getScorerPipeline(ScorerUtils.BASE_CONF)
+                .scoring(sceneFeatureMap, userFeatureMap, rankItems);
+        log.info("mergeAndRankRovRecallNew rovRecallScore={}", JSONUtils.toJson(rovRecallScore));
+        JSONObject obj = new JSONObject();
+        obj.put("name", "user_key_in_model_is_not_null");
+        log.info(obj.toString());
+        return rovRecallScore;
+    }
+
+    private Map<String, String> getSceneFeature(RankParam param) {
+        Map<String, String> sceneFeatureMap = new HashMap<>();
+        String provinceCn = param.getProvince();
+        provinceCn = provinceCn.replaceAll("省$", "");
+        sceneFeatureMap.put("ctx_region", provinceCn);
+        String city = param.getCity();
+        if ("台北市".equals(city) |
+                "高雄市".equals(city) |
+                "台中市".equals(city) |
+                "桃园市".equals(city) |
+                "新北市".equals(city) |
+                "台南市".equals(city) |
+                "基隆市".equals(city) |
+                "吉林市".equals(city) |
+                "新竹市".equals(city) |
+                "嘉义市".equals(city)
+        ){
+            ;
+        }else{
+            city = city.replaceAll("市$", "");
+        }
+        sceneFeatureMap.put("ctx_city", city);
+
+        Calendar calendar = Calendar.getInstance();
+        sceneFeatureMap.put("ctx_week", (calendar.get(Calendar.DAY_OF_WEEK) + 6) % 7 + "");
+        sceneFeatureMap.put("ctx_hour", new SimpleDateFormat("HH").format(calendar.getTime()));
+
+        return sceneFeatureMap;
+    }
+
+}