Browse Source

项目名称抽取

zhangjinkun 6 năm trước cách đây
mục cha
commit
35f0a7318b
2 tập tin đã thay đổi với 19 bổ sung12 xóa
  1. 12 10
      src/jy/extract/extract.go
  2. 7 2
      src/res/fieldscore.json

+ 12 - 10
src/jy/extract/extract.go

@@ -392,17 +392,19 @@ func (e *ExtractTask) ExtractDetail(j *ju.Job) {
 				// log.Debug("抽取-后置规则", tmp)
 
 				//项目名称未能抽取到,标题来凑
-				if vc.Field == "projectname" && vc.ExtFrom == "title" {
-					isextitle := true
-					for _, v := range j.Result[vc.Field] {
-						if len([]rune(qu.ObjToString(v.Value))) > 5 {
-							isextitle = false
-							break
+				if vc.Field == "projectname" {
+					if vc.ExtFrom == "title" {
+						isextitle := true
+						for _, v := range j.Result[vc.Field] {
+							if len([]rune(qu.ObjToString(v.Value))) > 5 {
+								isextitle = false
+								break
+							}
+						}
+						if isextitle { //标题加入选举
+							field := &ju.ExtField{Field: vc.Field, Code: vc.Id + "_title", RuleText: "title", Type: "title", MatchType: "title", ExtFrom: vc.ExtFrom, SourceValue: j.Title, Value: j.Title}
+							j.Result[vc.Field] = append(j.Result[vc.Field], field)
 						}
-					}
-					if isextitle { //标题加入选举
-						field := &ju.ExtField{Field: vc.Field, Code: vc.Id + "_title", RuleText: "title", Type: "title", MatchType: "title", ExtFrom: vc.ExtFrom, SourceValue: j.Title, Value: j.Title}
-						j.Result[vc.Field] = append(j.Result[vc.Field], field)
 					}
 					for i := 0; i < 3; i++ {
 						for _, v := range vc.RuleBacks {

+ 7 - 2
src/res/fieldscore.json

@@ -64,6 +64,11 @@
                 "describe": "以*结尾",
                 "regstr": ".{2,100}(的|招标|公示|公告|谈判|公开|通知|采购文件|交易中心)$",
                 "score": -5
+            },
+            {
+                "describe": "包含词",
+                "regstr": "(万元|本项目)",
+                "score": -10
             }
         ],
         "length": [
@@ -103,14 +108,14 @@
                 "describe": "[gt,lte,score]",
                 "range": [
                     35,
-                    45,
+                    60,
                     1
                 ]
             },
             {
                 "describe": "[gt,∞,score]",
                 "range": [
-                    45,
+                    60,
                     -1,
                     -2
                 ]