Commit
•
ed10c8b
1
Parent(s):
6401527
Update README.md
Browse files
README.md
CHANGED
|
@@ -9,6 +9,38 @@ One of the key aspect of this dataset which makes it unique is that it provides
|
|
| 9 |
|
| 10 |
## Dataset Structure
|
| 11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
|
| 13 |
### Data Fields
|
| 14 |
|
|
|
|
| 9 |
|
| 10 |
## Dataset Structure
|
| 11 |
|
| 12 |
+
## Dataset Statistics
|
| 13 |
+
Table 1: Statistics on the length of the abstractive keyphrases for Train, Test and Validation splits of Inspec dataset.
|
| 14 |
+
|
| 15 |
+
| | Score on Train split(%) | Score on Test split(%) | Score on Valid. split(%) |
|
| 16 |
+
|:---------------:|:-----------------------:|:----------------------:|:------------------------:|
|
| 17 |
+
| Single token | 9.0 | 9.5 | 10.1 |
|
| 18 |
+
| Two tokens | 50.4 | 48.2 | 45.7 |
|
| 19 |
+
| Three tokens | 27.6 | 28.6 | 29.8 |
|
| 20 |
+
| Four tokens | 9.3 | 10.3 | 10.3 |
|
| 21 |
+
| Five tokens | 2.4 | 2.0 | 3.2 |
|
| 22 |
+
| Six tokens | 0.9 | 1.2 | 0.7 |
|
| 23 |
+
| Seven tokens | 0.3 | 0.2 | 0.2 |
|
| 24 |
+
| Eight tokens | 0.1 | 0 | 0.1 |
|
| 25 |
+
| Nine tokens | 0 | 0.1 | 0 |
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
Table 2: Set of Analyses on Train, Validation, and Test Splits.
|
| 29 |
+
|
| 30 |
+
| Type of Analysis | Result(Train split) | Result(Test split) | Result(Valid. split) |
|
| 31 |
+
|:----------------------------------------------:|:------------------------------:|:------------------------------:|:------------------------------:|
|
| 32 |
+
| Annotator Type | Professional Indexer | Professional Indexer | Professional Indexer |
|
| 33 |
+
| Document Type | Abstracts from Inspec Database | Abstracts from Inspec Database | Abstracts from Inspec Database |
|
| 34 |
+
| Document Number | 1000 | 500 | 500 |
|
| 35 |
+
| Avg. Document length | 141.5 | 134.6 | 132.6 |
|
| 36 |
+
| Max Document length | 557 | 384 | 330 |
|
| 37 |
+
| Max # of abstractive keyphrases per document | 17 | 20 | 14 |
|
| 38 |
+
| Min # of abstractive keyphrases per document | 0 | 0 | 0 |
|
| 39 |
+
| Avg. # of abstractive keyphrases per document | 3.39 | 3.26 | 3.12 |
|
| 40 |
+
| Max # of extractive keyphrases per document | 24 | 27 | 22 |
|
| 41 |
+
| Min # of extractive keyphrases per document | 0 | 0 | 0 |
|
| 42 |
+
| Avg. # of extractive keyphrases per document | 6.39 | 6.56 | 5.95 |
|
| 43 |
+
|
| 44 |
|
| 45 |
### Data Fields
|
| 46 |
|