Update README.md
Browse files
README.md
CHANGED
@@ -1,74 +1,195 @@
|
|
1 |
-
---
|
2 |
-
dataset_info:
|
3 |
-
features:
|
4 |
-
- name: tokens
|
5 |
-
sequence: string
|
6 |
-
- name: ner_tags
|
7 |
-
sequence:
|
8 |
-
class_label:
|
9 |
-
names:
|
10 |
-
'0': O
|
11 |
-
'1': B-UoM
|
12 |
-
'2': I-UoM
|
13 |
-
'3': B-color
|
14 |
-
'4': I-color
|
15 |
-
'5': B-condition
|
16 |
-
'6': I-condition
|
17 |
-
'7': B-content
|
18 |
-
'8': I-content
|
19 |
-
'9': B-core_product_type
|
20 |
-
'10': I-core_product_type
|
21 |
-
'11': B-creator
|
22 |
-
'12': I-creator
|
23 |
-
'13': B-department
|
24 |
-
'14': I-department
|
25 |
-
'15': B-material
|
26 |
-
'16': I-material
|
27 |
-
'17': B-modifier
|
28 |
-
'18': I-modifier
|
29 |
-
'19': B-occasion
|
30 |
-
'20': I-occasion
|
31 |
-
'21': B-origin
|
32 |
-
'22': I-origin
|
33 |
-
'23': B-price
|
34 |
-
'24': I-price
|
35 |
-
'25': B-product_name
|
36 |
-
'26': I-product_name
|
37 |
-
'27': B-product_number
|
38 |
-
'28': I-product_number
|
39 |
-
'29': B-quantity
|
40 |
-
'30': I-quantity
|
41 |
-
'31': B-shape
|
42 |
-
'32': I-shape
|
43 |
-
'33': B-time
|
44 |
-
'34': I-time
|
45 |
-
splits:
|
46 |
-
- name: train
|
47 |
-
num_bytes: 553523
|
48 |
-
num_examples: 7841
|
49 |
-
- name: test
|
50 |
-
num_bytes: 70308
|
51 |
-
num_examples: 993
|
52 |
-
- name: validation
|
53 |
-
num_bytes: 61109
|
54 |
-
num_examples: 871
|
55 |
-
download_size: 242711
|
56 |
-
dataset_size: 684940
|
57 |
-
configs:
|
58 |
-
- config_name: default
|
59 |
-
data_files:
|
60 |
-
- split: train
|
61 |
-
path: data/train-*
|
62 |
-
- split: test
|
63 |
-
path: data/test-*
|
64 |
-
- split: validation
|
65 |
-
path: data/validation-*
|
66 |
-
license: cc-by-4.0
|
67 |
-
task_categories:
|
68 |
-
- token-classification
|
69 |
-
language:
|
70 |
-
- en
|
71 |
-
pretty_name: QueryNER
|
72 |
-
size_categories:
|
73 |
-
- 1K<n<10K
|
74 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
dataset_info:
|
3 |
+
features:
|
4 |
+
- name: tokens
|
5 |
+
sequence: string
|
6 |
+
- name: ner_tags
|
7 |
+
sequence:
|
8 |
+
class_label:
|
9 |
+
names:
|
10 |
+
'0': O
|
11 |
+
'1': B-UoM
|
12 |
+
'2': I-UoM
|
13 |
+
'3': B-color
|
14 |
+
'4': I-color
|
15 |
+
'5': B-condition
|
16 |
+
'6': I-condition
|
17 |
+
'7': B-content
|
18 |
+
'8': I-content
|
19 |
+
'9': B-core_product_type
|
20 |
+
'10': I-core_product_type
|
21 |
+
'11': B-creator
|
22 |
+
'12': I-creator
|
23 |
+
'13': B-department
|
24 |
+
'14': I-department
|
25 |
+
'15': B-material
|
26 |
+
'16': I-material
|
27 |
+
'17': B-modifier
|
28 |
+
'18': I-modifier
|
29 |
+
'19': B-occasion
|
30 |
+
'20': I-occasion
|
31 |
+
'21': B-origin
|
32 |
+
'22': I-origin
|
33 |
+
'23': B-price
|
34 |
+
'24': I-price
|
35 |
+
'25': B-product_name
|
36 |
+
'26': I-product_name
|
37 |
+
'27': B-product_number
|
38 |
+
'28': I-product_number
|
39 |
+
'29': B-quantity
|
40 |
+
'30': I-quantity
|
41 |
+
'31': B-shape
|
42 |
+
'32': I-shape
|
43 |
+
'33': B-time
|
44 |
+
'34': I-time
|
45 |
+
splits:
|
46 |
+
- name: train
|
47 |
+
num_bytes: 553523
|
48 |
+
num_examples: 7841
|
49 |
+
- name: test
|
50 |
+
num_bytes: 70308
|
51 |
+
num_examples: 993
|
52 |
+
- name: validation
|
53 |
+
num_bytes: 61109
|
54 |
+
num_examples: 871
|
55 |
+
download_size: 242711
|
56 |
+
dataset_size: 684940
|
57 |
+
configs:
|
58 |
+
- config_name: default
|
59 |
+
data_files:
|
60 |
+
- split: train
|
61 |
+
path: data/train-*
|
62 |
+
- split: test
|
63 |
+
path: data/test-*
|
64 |
+
- split: validation
|
65 |
+
path: data/validation-*
|
66 |
+
license: cc-by-4.0
|
67 |
+
task_categories:
|
68 |
+
- token-classification
|
69 |
+
language:
|
70 |
+
- en
|
71 |
+
pretty_name: QueryNER
|
72 |
+
size_categories:
|
73 |
+
- 1K<n<10K
|
74 |
+
---
|
75 |
+
# Dataset Card for QueryNER
|
76 |
+
|
77 |
+
QueryNER is a sequence labeling dataset for e-commerce query segmentation.
|
78 |
+
It has 17 different entity types. QueryNER covers nearly the entire query rather than just certain key aspects that may be covered by other aspect-value extraction systems.
|
79 |
+
|
80 |
+
|
81 |
+
## Dataset Details
|
82 |
+
|
83 |
+
### Dataset Description
|
84 |
+
|
85 |
+
QueryNER is a manually-annotated dataset and accompanying model for e-commerce query segmentation. Prior work in sequence labeling for e-commerce has largely addressed aspect-value extraction which focuses
|
86 |
+
on extracting portions of a product title or query for narrowly defined aspects. Our work instead focuses on the goal
|
87 |
+
of dividing a query into meaningful chunks with broadly applicable types.
|
88 |
+
QueryNER has 17 different entity types.
|
89 |
+
|
90 |
+
|
91 |
+
- **Curated by:** BLT Lab
|
92 |
+
- **Language(s) (NLP):** English
|
93 |
+
- **License:** CC-BY 4.0
|
94 |
+
|
95 |
+
### Dataset Sources
|
96 |
+
|
97 |
+
QueryNER is annotation on a subsection of Amazon's (ESCI Shopping Queries dataset)[https://github.com/amazon-science/esci-data].
|
98 |
+
|
99 |
+
|
100 |
+
- **Repository:**
|
101 |
+
- **Paper [optional]:** [More Information Needed]
|
102 |
+
- **Demo [optional]:** [More Information Needed]
|
103 |
+
|
104 |
+
## Uses
|
105 |
+
|
106 |
+
QueryNER is intended to be used for segmentation of e-commerce queries in English.
|
107 |
+
|
108 |
+
### Direct Use
|
109 |
+
|
110 |
+
QueryNER can be used for research on e-commerce query segmentation.
|
111 |
+
It may also be used for e-commerce query segmentation for use in further downstream systems; however, we caution users that while the ontology is broadly applicable, using models trained on only this small public release may have suboptimal performance especially on out of domain data.
|
112 |
+
|
113 |
+
### Out-of-Scope Use
|
114 |
+
|
115 |
+
Users would likely experience poor segmentation performance on data outside of the e-commerce domain.
|
116 |
+
Because the dataset is on the smaller side, additional annotated data on additional data using the QueryNER ontology
|
117 |
+
may be necessary to get better performance on other datasets.
|
118 |
+
|
119 |
+
|
120 |
+
## Dataset Structure
|
121 |
+
|
122 |
+
The dataset includes the query tokens and their tags.
|
123 |
+
|
124 |
+
|
125 |
+
## Dataset Creation
|
126 |
+
See paper.
|
127 |
+
|
128 |
+
### Curation Rationale
|
129 |
+
|
130 |
+
The dataset was created for research and for downstream applications for e-commerce search systems to make use of segmented queries.
|
131 |
+
|
132 |
+
|
133 |
+
### Source Data
|
134 |
+
|
135 |
+
The source data is from the Shopping Queries ESCI dataset.
|
136 |
+
(https://github.com/amazon-science/esci-data)[https://github.com/amazon-science/esci-data]
|
137 |
+
```
|
138 |
+
@article{reddy2022shopping,
|
139 |
+
title={Shopping Queries Dataset: A Large-Scale {ESCI} Benchmark for Improving Product Search},
|
140 |
+
author={Chandan K. Reddy and Lluís Màrquez and Fran Valero and Nikhil Rao and Hugo Zaragoza and Sambaran Bandyopadhyay and Arnab Biswas and Anlu Xing and Karthik Subbian},
|
141 |
+
year={2022},
|
142 |
+
eprint={2206.06588},
|
143 |
+
archivePrefix={arXiv}
|
144 |
+
}
|
145 |
+
```
|
146 |
+
|
147 |
+
#### Data Collection and Processing
|
148 |
+
|
149 |
+
See paper
|
150 |
+
|
151 |
+
|
152 |
+
#### Who are the source data producers?
|
153 |
+
|
154 |
+
See source data repo and paper.
|
155 |
+
|
156 |
+
|
157 |
+
### Annotations
|
158 |
+
|
159 |
+
#### Annotation process
|
160 |
+
|
161 |
+
See paper for details.
|
162 |
+
|
163 |
+
#### Who are the annotators?
|
164 |
+
|
165 |
+
Annotators were contract workers and were paid a living wage.
|
166 |
+
|
167 |
+
#### Personal and Sensitive Information
|
168 |
+
|
169 |
+
The dataset is just user e-commerce queries and should not contain any sensitive information.
|
170 |
+
|
171 |
+
|
172 |
+
## Bias, Risks, and Limitations
|
173 |
+
|
174 |
+
The dataset is English only for now.
|
175 |
+
Bias may be toward e-commerce queries of the source data.
|
176 |
+
There may also be annotator bias since the dataset is annotated by a single annotator for the training set and three annotators and an adjudicator for the development and test sets.
|
177 |
+
|
178 |
+
|
179 |
+
## Citation [optional]
|
180 |
+
|
181 |
+
To appear at LREC-COLING 2024.
|
182 |
+
|
183 |
+
**BibTeX:**
|
184 |
+
|
185 |
+
Coming soon
|
186 |
+
|
187 |
+
|
188 |
+
|
189 |
+
## Dataset Card Authors
|
190 |
+
|
191 |
+
Chester Palen-Michel @cpalenmichel
|
192 |
+
|
193 |
+
## Dataset Card Contact
|
194 |
+
|
195 |
+
Chester Palen-Michel @cpalenmichel
|