topdog/supersonic
1
0
Fork 0
mirror of https://github.com/tencentmusic/supersonic.git synced 2025-12-12 20:51:48 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

(fix)(chat) Fix the issue of mapping in the map (#1825)

Browse Source
This commit is contained in:
lexluo09
2024-10-19 13:22:39 +08:00
committed by GitHub
parent bb07fd77cf
commit 2250289c94
5 changed files with 25 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
headless/chat/src/main/java/com/tencent/supersonic/headless/chat/knowledge/SearchService.java
View File

@@ -58,9 +58,8 @@ public class SearchService {
return new HanlpMapResult(name, entry.getValue(), key, similarity);
}).sorted((a, b) -> -(b.getName().length() - a.getName().length()))
.collect(Collectors.toList());
hanlpMapResults = transformAndFilterByDataSet(hanlpMapResults, modelIdToDataSetIds,
detectDataSetIds, limit);
return hanlpMapResults;
return transformAndFilterByDataSet(hanlpMapResults, modelIdToDataSetIds, detectDataSetIds,
limit);
}
/**
@@ -71,15 +70,14 @@ public class SearchService {
*/
public static List<HanlpMapResult> suffixSearch(String key, int limit,
Map<Long, List<Long>> modelIdToDataSetIds, Set<Long> detectDataSetIds) {
String reverseDetectSegment = StringUtils.reverse(key);
return suffixSearch(reverseDetectSegment, limit, suffixTrie, modelIdToDataSetIds,
detectDataSetIds);
return suffixSearch(key, limit, suffixTrie, modelIdToDataSetIds, detectDataSetIds);
}
public static List<HanlpMapResult> suffixSearch(String key, int limit,
BinTrie<List<String>> binTrie, Map<Long, List<Long>> modelIdToDataSetIds,
Set<Long> detectDataSetIds) {
Set<Map.Entry<String, List<String>>> result = search(key, binTrie);
String reverseDetectSegment = StringUtils.reverse(key);
Set<Map.Entry<String, List<String>>> result = search(reverseDetectSegment, binTrie);
List<HanlpMapResult> hanlpMapResults = result.stream().map(entry -> {
String name = entry.getKey().replace("#", " ");
List<String> natures = entry.getValue().stream()

6
headless/chat/src/main/java/com/tencent/supersonic/headless/chat/mapper/MapFilter.java
View File

@@ -77,9 +77,9 @@ public class MapFilter {
for (Map.Entry<Long, List<SchemaElementMatch>> entry : dataSetElementMatches.entrySet()) {
List<SchemaElementMatch> value = entry.getValue();
if (!CollectionUtils.isEmpty(value)) {
value.removeIf(schemaElementMatch -> StringUtils
.length(schemaElementMatch.getDetectWord()) <= 2
&& schemaElementMatch.getSimilarity() < 1);
value.removeIf(
schemaElementMatch -> StringUtils.length(schemaElementMatch.getWord()) <= 2
&& schemaElementMatch.getSimilarity() < 1);
}
}
}

2
headless/chat/src/main/java/com/tencent/supersonic/headless/chat/mapper/MapperConfig.java
View File

@@ -15,7 +15,7 @@ public class MapperConfig extends ParameterConfig {
"number", "Mapper相关配置");
public static final Parameter MAPPER_NAME_THRESHOLD =
new Parameter("s2.mapper.name.threshold", "0.5", "指标名、维度名文本相似度阈值",
new Parameter("s2.mapper.name.threshold", "0.3", "指标名、维度名文本相似度阈值",
"文本片段和匹配到的指标、维度名计算出来的编辑距离阈值, 若超出该阈值, 则舍弃", "number", "Mapper相关配置");
public static final Parameter MAPPER_NAME_THRESHOLD_MIN =