File size: 1,000 Bytes
e68ceba
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
---
pretty_name: UkrainIA canonicalization
task_categories:
  - Open Knowledge Graph Canonicalization
tags:
  - entity-resolution
  - knowledge-graphs
  - deduplication
license: mit
---

# Entity Duplicate Candidates

A dataset of candidate duplicate entity pairs with features and labels, sorted by `original`.

## Columns
- **original** *(string)*: Source entity surface form.
- **candidate** *(string)*: Candidate entity surface form.
- **levenshtein** *(float/int)*: Levenshtein distance (or normalized score if applicable).
- **normalized** *(string/bool)*: Normalized surface form or flag (as in your source).
- **hashtag** *(string/bool)*: Hashtag feature (as in your source).
- **cossim_score** *(float)*: Cosine similarity between sentence embeddings.
- **len_shortest_path** *(int)*: Length of shortest path between entities in the KG (if any).
- **nb_shortest_path** *(int)*: Number of shortest paths.
- **is_duplicate** *(bool)*: Ground-truth label (True if the pair is a duplicate).

```