Cross-Lingual Author Disambiguation at Scale: Difference between revisions

Content deleted Content added

Inline

@@ Line 7: / Line 7: @@
             "statements": {
                 "Title": {
-                    "type": "text",
+                    "propertyType": "text",
                     "value": [
                         "Cross-Lingual Author Disambiguation at Scale"
@@ Line 13: / Line 13: @@
                 },
                 "Year": {
-                    "type": "number",
+                    "propertyType": "number",
                     "value": 2023
                 },
                 "Venue": {
-                    "type": "text",
+                    "propertyType": "text",
                     "value": [
                         "ECIR Proceedings"
@@ Line 23: / Line 23: @@
                 },
                 "Authors": {
-                    "type": "relation",
+                    "propertyType": "relation",
                     "value": [
                         {
@@ Line 36: / Line 36: @@
                 },
                 "Project": {
-                    "type": "relation",
+                    "propertyType": "relation",
                     "value": [
                         {
@@ Line 45: / Line 45: @@
                 },
                 "URL": {
-                    "type": "url",
+                    "propertyType": "url",
                     "value": [
                         "https://example.org/publications/cross-lingual-disambiguation"

Latest revision as of 22:06, 27 July 2026

"Cross-Lingual Author Disambiguation at Scale" presents methods for resolving author identities across bibliographic datasets that span multiple languages and transliteration conventions. The paper combines name normalisation, co-author network signals, and topic-aware embeddings to disambiguate authors at the scale of national catalogue exports.

The work feeds into the Multilingual NER Toolkit project, where the disambiguation pipeline integrates with upstream entity recognition for historical European texts.