introduce unknown concept and handling

src/tagpack/db/concepts.yaml

@@ -20,6 +20,14 @@ concept:
     Anything that describes an actor in the Darknet- or Cryptoasset Ecosystems.
   narrower: [entity, behaviour]
 
+unknown:
+  id: unknown
+  type: concept
+  prefLabel: Unknown
+  description: >
+    Special category representing a tag with unknown categorization
+  broader: concept
+
 behaviour:
   id: behaviour
   type: concept

src/tagstore/algorithms/tag_digest.py

@@ -148,6 +148,12 @@ def add_tag_data(t, tags_count, total_words):
                     concepts_counter.add(x, weight=conf * _get_concept_weight(x))
 
                     ls["concepts"].add(x)
+            else:
+                # tags without categorization are added to unknown category in wordcloud
+                x = "unknown"
+                concepts_counter.add(x, weight=conf * _get_concept_weight(x))
+
+                ls["concepts"].add(x)
 
             ls["cnt"] += 1
             ls["lbl"] = t.label
@@ -181,7 +187,7 @@ def add_tag_data(t, tags_count, total_words):
             ].capitalize()
 
     # get broad category
-    broad_concept = "user"
+    broad_concept = "entity"
     if len(concepts_counter) > 0:
         broad_concept = _map_concept_to_broad_concept(
             concepts_counter.most_common(1, weighted=True)[0][0]

src/tagstore/db/queries.py

@@ -200,11 +200,13 @@ class TagPublic(BaseModel):
     @computed_field
     @property
     def concepts(self) -> List[str]:
-        return (
-            [self.primary_concept] + self.additional_concepts
-            if (self.primary_concept)
-            else self.additional_concepts
-        )
+        return dict.fromkeys(
+            (
+                [self.primary_concept] + self.additional_concepts
+                if (self.primary_concept)
+                else self.additional_concepts
+            )
+        ).keys()
 
     @classmethod
     def fromDB(cls, t: Tag, tp: TagPack, inherited_from=None) -> "TagPublic":