espejelomar commited on
Commit
aa3d54a
1 Parent(s): ad8fd84

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +35 -21
README.md CHANGED
@@ -4,6 +4,11 @@ language:
4
  - en
5
  paperswithcode_id: embedding-data/WikiAnswers
6
  pretty_name: WikiAnswers
 
 
 
 
 
7
  ---
8
 
9
  # Dataset Card for "WikiAnswers"
@@ -43,15 +48,38 @@ pretty_name: WikiAnswers
43
  The WikiAnswers corpus contains clusters of questions tagged by WikiAnswers users as paraphrases.
44
  Each cluster optionally contains an answer provided by WikiAnswers users. There are 30,370,994 clusters containing an average of 25 questions per cluster. 3,386,256 (11%) of the clusters have an answer.
45
 
46
- ### Supported Tasks and Leaderboards
47
-
48
- [More Information Needed](https://github.com/afader/oqa#wikianswers-corpus)
49
-
50
  ### Languages
51
-
52
- [More Information Needed](https://github.com/afader/oqa#wikianswers-corpus)
53
-
54
  ## Dataset Structure
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
55
 
56
  ### Data Instances
57
 
@@ -59,19 +87,6 @@ Each cluster optionally contains an answer provided by WikiAnswers users. There
59
 
60
  ### Data Splits
61
 
62
- The data can be downloaded from: [http://knowitall.cs.washington.edu/oqa/data/wikianswers/](http://knowitall.cs.washington.edu/oqa/data/wikianswers/).
63
- The corpus is split into 40 gzip-compressed files. The total compressed filesize is 8GB; the total decompressed filesize is 40GB.
64
- Each file contains one cluster per line. Each cluster is a tab-separated list of questions and answers.
65
- Questions are prefixed by q: and answers are prefixed by a:. Here is an example cluster (tabs replaced with newlines):
66
-
67
- ```
68
- q:How many muslims make up indias 1 billion population?
69
- q:How many of india's population are muslim?
70
- q:How many populations of muslims in india?
71
- q:What is population of muslims in india?
72
- a:Over 160 million Muslims per Pew Forum Study as of October 2009.
73
-
74
- ```
75
 
76
  ## Dataset Creation
77
 
@@ -142,5 +157,4 @@ a:Over 160 million Muslims per Pew Forum Study as of October 2009.
142
 
143
  ### Contributions
144
 
145
- Thanks to [Anthony Fader](https://dl.acm.org/profile/81324489111), [Luke Zettlemoyer](https://dl.acm.org/profile/81100527621), [Oren Etzioni](https://dl.acm.org/profile/99658633129) for adding this dataset.
146
 
4
  - en
5
  paperswithcode_id: embedding-data/WikiAnswers
6
  pretty_name: WikiAnswers
7
+ task_categories:
8
+ - sentence-similarity
9
+ - paraphrase-mining
10
+ task_ids:
11
+ - semantic-similarity-classification
12
  ---
13
 
14
  # Dataset Card for "WikiAnswers"
48
  The WikiAnswers corpus contains clusters of questions tagged by WikiAnswers users as paraphrases.
49
  Each cluster optionally contains an answer provided by WikiAnswers users. There are 30,370,994 clusters containing an average of 25 questions per cluster. 3,386,256 (11%) of the clusters have an answer.
50
 
51
+ ### Supported Tasks
52
+ - [Sentence Transformers](https://huggingface.co/sentence-transformers) training; useful for semantic search and sentence similarity.
 
 
53
  ### Languages
54
+ - English.
 
 
55
  ## Dataset Structure
56
+ Each example in the dataset contains 25 equivalent sentences and is formatted as a dictionary with the key "set" and a list with the sentences as "value".
57
+ ```
58
+ {"set": [sentence_1, sentence_2, ..., sentence_25]}
59
+ {"set": [sentence_1, sentence_2, ..., sentence_25]}
60
+ ...
61
+ {"set": [sentence_1, sentence_2, ..., sentence_25]}
62
+ ```
63
+ This dataset is useful for training Sentence Transformers models. Refer to the following post on how to train models using similar sentences.
64
+ ### Usage Example
65
+ Install the 🤗 Datasets library with `pip install datasets` and load the dataset from the Hub with:
66
+ ```python
67
+ from datasets import load_dataset
68
+ dataset = load_dataset("embedding-data/WikiAnswers")
69
+ ```
70
+ The dataset is loaded as a `DatasetDict` and has the format for `N` examples:
71
+ ```python
72
+ DatasetDict({
73
+ train: Dataset({
74
+ features: ['set'],
75
+ num_rows: N
76
+ })
77
+ })
78
+ ```
79
+ Review an example `i` with:
80
+ ```python
81
+ dataset["train"][i]["set"]
82
+ ```
83
 
84
  ### Data Instances
85
 
87
 
88
  ### Data Splits
89
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
 
91
  ## Dataset Creation
92
 
157
 
158
  ### Contributions
159
 
 
160