Datasets:

Modalities:
Text
Languages:
English
Libraries:
Datasets
License:
asahi417 commited on
Commit
d8cb616
1 Parent(s): 4eb79ee

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -269,14 +269,14 @@ The number of triples in each split is summarized in the table below.
269
 
270
 
271
  ### Filtering to Remove Noise
272
- We apply filtering to keep triples with alpha-numeric subject and object, as well as triples with at least either of subject or object is a named-entity.
 
273
  After the filtering, we manually remove too vague and noisy predicate, and unify same predicates with different names (see the annotation [here](https://huggingface.co/datasets/relbert/t_rex/raw/main/predicate_manual_check.csv)).
274
 
275
- | Dataset | `raw` | `filter` | `filter_unified` |
276
- |:----------|----------:|----------:|-----------------:|
277
- | Triples | 941,663 | 583,333 | 432,795 |
278
- | Predicate | 931 | 659 | 247 |
279
- | Entity | 270,801 | 197,163 | 149,172 |
280
 
281
  ### Filtering to Purify the Dataset
282
  We reduce the size of the dataset by applying filtering based on the number of predicates and entities in the triples.
 
269
 
270
 
271
  ### Filtering to Remove Noise
272
+ We apply filtering to keep triples with named-entities in either of head or tail (`named-entity filter`).
273
+ Then, we remove predicates if they have less than three triples (`rare-predicate filter`).
274
  After the filtering, we manually remove too vague and noisy predicate, and unify same predicates with different names (see the annotation [here](https://huggingface.co/datasets/relbert/t_rex/raw/main/predicate_manual_check.csv)).
275
 
276
+ | Dataset | `raw` | `named-entity filter` | `rare-predicate` |
277
+ |:----------|-----------:|-----------------------:|-----------------:|
278
+ | Triples | 20,877,472 | 12,561,573 | 12,561,250 |
279
+ | Predicate | 1,616 | 1,470 | 1,237 |
 
280
 
281
  ### Filtering to Purify the Dataset
282
  We reduce the size of the dataset by applying filtering based on the number of predicates and entities in the triples.