microsoft/msr_sqa
收藏资源简介:
--- annotations_creators: - crowdsourced language_creators: - found language: - en license: - ms-pl multilinguality: - monolingual size_categories: - 10K<n<100K source_datasets: - original task_categories: - question-answering task_ids: - extractive-qa paperswithcode_id: null pretty_name: Microsoft Research Sequential Question Answering dataset_info: features: - name: id dtype: string - name: annotator dtype: int32 - name: position dtype: int32 - name: question dtype: string - name: question_and_history sequence: string - name: table_file dtype: string - name: table_header sequence: string - name: table_data sequence: sequence: string - name: answer_coordinates sequence: - name: row_index dtype: int32 - name: column_index dtype: int32 - name: answer_text sequence: string splits: - name: train num_bytes: 19732499 num_examples: 12276 - name: validation num_bytes: 3738331 num_examples: 2265 - name: test num_bytes: 5105873 num_examples: 3012 download_size: 4796932 dataset_size: 28576703 --- # Dataset Card for Microsoft Research Sequential Question Answering ## Table of Contents - [Dataset Description](#dataset-description) - [Dataset Summary](#dataset-summary) - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards) - [Languages](#languages) - [Dataset Structure](#dataset-structure) - [Data Instances](#data-instances) - [Data Fields](#data-fields) - [Data Splits](#data-splits) - [Dataset Creation](#dataset-creation) - [Curation Rationale](#curation-rationale) - [Source Data](#source-data) - [Annotations](#annotations) - [Personal and Sensitive Information](#personal-and-sensitive-information) - [Considerations for Using the Data](#considerations-for-using-the-data) - [Social Impact of Dataset](#social-impact-of-dataset) - [Discussion of Biases](#discussion-of-biases) - [Other Known Limitations](#other-known-limitations) - [Additional Information](#additional-information) - [Dataset Curators](#dataset-curators) - [Licensing Information](#licensing-information) - [Citation Information](#citation-information) - [Contributions](#contributions) ## Dataset Description - **Homepage:** [Microsoft Research Sequential Question Answering (SQA) Dataset](https://msropendata.com/datasets/b25190ed-0f59-47b1-9211-5962858142c2) - **Repository:** - **Paper:** [https://www.microsoft.com/en-us/research/wp-content/uploads/2017/05/acl17-dynsp.pdf](https://www.microsoft.com/en-us/research/wp-content/uploads/2017/05/acl17-dynsp.pdf) - **Leaderboard:** - **Point of Contact:** - Scott Wen-tau Yih scottyih@microsoft.com - Mohit Iyyer m.iyyer@gmail.com - Ming-Wei Chang minchang@microsoft.com ### Dataset Summary Recent work in semantic parsing for question answering has focused on long and complicated questions, many of which would seem unnatural if asked in a normal conversation between two humans. In an effort to explore a conversational QA setting, we present a more realistic task: answering sequences of simple but inter-related questions. We created SQA by asking crowdsourced workers to decompose 2,022 questions from WikiTableQuestions (WTQ)*, which contains highly-compositional questions about tables from Wikipedia. We had three workers decompose each WTQ question, resulting in a dataset of 6,066 sequences that contain 17,553 questions in total. Each question is also associated with answers in the form of cell locations in the tables. - Panupong Pasupat, Percy Liang. "Compositional Semantic Parsing on Semi-Structured Tables" ACL-2015. [http://www-nlp.stanford.edu/software/sempre/wikitable/](http://www-nlp.stanford.edu/software/sempre/wikitable/) ### Supported Tasks and Leaderboards [More Information Needed] ### Languages English (`en`). ## Dataset Structure ### Data Instances ``` {'id': 'nt-639', 'annotator': 0, 'position': 0, 'question': 'where are the players from?', 'table_file': 'table_csv/203_149.csv', 'table_header': ['Pick', 'Player', 'Team', 'Position', 'School'], 'table_data': [['1', 'Ben McDonald', 'Baltimore Orioles', 'RHP', 'Louisiana State University'], ['2', 'Tyler Houston', 'Atlanta Braves', 'C', '"Valley HS (Las Vegas', ' NV)"'], ['3', 'Roger Salkeld', 'Seattle Mariners', 'RHP', 'Saugus (CA) HS'], ['4', 'Jeff Jackson', 'Philadelphia Phillies', 'OF', '"Simeon HS (Chicago', ' IL)"'], ['5', 'Donald Harris', 'Texas Rangers', 'OF', 'Texas Tech University'], ['6', 'Paul Coleman', 'Saint Louis Cardinals', 'OF', 'Frankston (TX) HS'], ['7', 'Frank Thomas', 'Chicago White Sox', '1B', 'Auburn University'], ['8', 'Earl Cunningham', 'Chicago Cubs', 'OF', 'Lancaster (SC) HS'], ['9', 'Kyle Abbott', 'California Angels', 'LHP', 'Long Beach State University'], ['10', 'Charles Johnson', 'Montreal Expos', 'C', '"Westwood HS (Fort Pierce', ' FL)"'], ['11', 'Calvin Murray', 'Cleveland Indians', '3B', '"W.T. White High School (Dallas', ' TX)"'], ['12', 'Jeff Juden', 'Houston Astros', 'RHP', 'Salem (MA) HS'], ['13', 'Brent Mayne', 'Kansas City Royals', 'C', 'Cal State Fullerton'], ['14', 'Steve Hosey', 'San Francisco Giants', 'OF', 'Fresno State University'], ['15', 'Kiki Jones', 'Los Angeles Dodgers', 'RHP', '"Hillsborough HS (Tampa', ' FL)"'], ['16', 'Greg Blosser', 'Boston Red Sox', 'OF', 'Sarasota (FL) HS'], ['17', 'Cal Eldred', 'Milwaukee Brewers', 'RHP', 'University of Iowa'], ['18', 'Willie Greene', 'Pittsburgh Pirates', 'SS', '"Jones County HS (Gray', ' GA)"'], ['19', 'Eddie Zosky', 'Toronto Blue Jays', 'SS', 'Fresno State University'], ['20', 'Scott Bryant', 'Cincinnati Reds', 'OF', 'University of Texas'], ['21', 'Greg Gohr', 'Detroit Tigers', 'RHP', 'Santa Clara University'], ['22', 'Tom Goodwin', 'Los Angeles Dodgers', 'OF', 'Fresno State University'], ['23', 'Mo Vaughn', 'Boston Red Sox', '1B', 'Seton Hall University'], ['24', 'Alan Zinter', 'New York Mets', 'C', 'University of Arizona'], ['25', 'Chuck Knoblauch', 'Minnesota Twins', '2B', 'Texas A&M University'], ['26', 'Scott Burrell', 'Seattle Mariners', 'RHP', 'Hamden (CT) HS']], 'answer_coordinates': {'row_index': [0, 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], 'column_index': [4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4]}, 'answer_text': ['Louisiana State University', 'Valley HS (Las Vegas, NV)', 'Saugus (CA) HS', 'Simeon HS (Chicago, IL)', 'Texas Tech University', 'Frankston (TX) HS', 'Auburn University', 'Lancaster (SC) HS', 'Long Beach State University', 'Westwood HS (Fort Pierce, FL)', 'W.T. White High School (Dallas, TX)', 'Salem (MA) HS', 'Cal State Fullerton', 'Fresno State University', 'Hillsborough HS (Tampa, FL)', 'Sarasota (FL) HS', 'University of Iowa', 'Jones County HS (Gray, GA)', 'Fresno State University', 'University of Texas', 'Santa Clara University', 'Fresno State University', 'Seton Hall University', 'University of Arizona', 'Texas A&M University', 'Hamden (CT) HS']} ``` ### Data Fields - `id` (`str`): question sequence id (the id is consistent with those in WTQ) - `annotator` (`int`): `0`, `1`, `2` (the 3 annotators who annotated the question intent) - `position` (`int`): the position of the question in the sequence - `question` (`str`): the question given by the annotator - `table_file` (`str`): the associated table - `table_header` (`List[str]`): a list of headers in the table - `table_data` (`List[List[str]]`): 2d array of data in the table - `answer_coordinates` (`List[Dict]`): the table cell coordinates of the answers (0-based, where 0 is the first row after the table header) - `row_index` - `column_index` - `answer_text` (`List[str]`): the content of the answer cells Note that some text fields may contain Tab or LF characters and thus start with quotes. It is recommended to use a CSV parser like the Python CSV package to process the data. ### Data Splits | | train | test | |-------------|------:|-----:| | N. examples | 14541 | 3012 | ## Dataset Creation ### Curation Rationale [More Information Needed] ### Source Data #### Initial Data Collection and Normalization [More Information Needed] #### Who are the source language producers? [More Information Needed] ### Annotations #### Annotation process [More Information Needed] #### Who are the annotators? [More Information Needed] ### Personal and Sensitive Information [More Information Needed] ## Considerations for Using the Data ### Social Impact of Dataset [More Information Needed] ### Discussion of Biases [More Information Needed] ### Other Known Limitations [More Information Needed] ## Additional Information ### Dataset Curators [More Information Needed] ### Licensing Information [Microsoft Research Data License Agreement](https://msropendata-web-api.azurewebsites.net/licenses/2f933be3-284d-500b-7ea3-2aa2fd0f1bb2/view). ### Citation Information ``` @inproceedings{iyyer-etal-2017-search, title = "Search-based Neural Structured Learning for Sequential Question Answering", author = "Iyyer, Mohit and Yih, Wen-tau and Chang, Ming-Wei", booktitle = "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = jul, year = "2017", address = "Vancouver, Canada", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/P17-1167", doi = "10.18653/v1/P17-1167", pages = "1821--1831", } ``` ### Contributions Thanks to [@mattbui](https://github.com/mattbui) for adding this dataset.
annotations_creators: - 众包(crowdsourced) language_creators: - 现有资源采集(found) language: - 英语(en) license: - 微软公共许可证(ms-pl, Microsoft Public License) multilinguality: - 单语言(monolingual) size_categories: - 10000 < 样本数 < 100000 source_datasets: - 原创数据集(original) task_categories: - 问答(question-answering) task_ids: - 抽取式问答(extractive QA) paperswithcode_id: 无 pretty_name: 微软研究院序列问答数据集(Microsoft Research Sequential Question Answering) dataset_info: features: - name: id dtype: string - name: annotator dtype: int32 - name: position dtype: int32 - name: question dtype: string - name: question_and_history sequence: string - name: table_file dtype: string - name: table_header sequence: string - name: table_data sequence: sequence: string - name: answer_coordinates sequence: - name: row_index dtype: int32 - name: column_index dtype: int32 - name: answer_text sequence: string splits: - name: train num_bytes: 19732499 num_examples: 12276 - name: validation num_bytes: 3738331 num_examples: 2265 - name: test num_bytes: 5105873 num_examples: 3012 download_size: 4796932 dataset_size: 28576703 # 微软研究院序列问答数据集(Microsoft Research Sequential Question Answering)数据集卡片 ## 目录 - [数据集描述](#dataset-description) - [数据集概述](#dataset-summary) - [支持任务与排行榜](#supported-tasks-and-leaderboards) - [语言](#languages) - [数据集结构](#dataset-structure) - [数据实例](#data-instances) - [数据字段](#data-fields) - [数据划分](#data-splits) - [数据集构建](#dataset-creation) - [构建初衷](#curation-rationale) - [源数据](#source-data) - [标注](#annotations) - [个人与敏感信息](#personal-and-sensitive-information) - [数据使用注意事项](#considerations-for-using-the-data) - [数据集的社会影响](#social-impact-of-dataset) - [偏差讨论](#discussion-of-biases) - [其他已知局限性](#other-known-limitations) - [附加信息](#additional-information) - [数据集维护者](#dataset-curators) - [许可信息](#licensing-information) - [引用信息](#citation-information) - [贡献](#contributions) ## 数据集描述 - **"主页"**:[微软研究院序列问答(SQA)数据集](https://msropendata.com/datasets/b25190ed-0f59-47b1-9211-5962858142c2) - **"代码仓库"**: - **"论文"**:[https://www.microsoft.com/en-us/research/wp-content/uploads/2017/05/acl17-dynsp.pdf](https://www.microsoft.com/en-us/research/wp-content/uploads/2017/05/acl17-dynsp.pdf) - **"排行榜"**: - **"联系方式"**: - Scott Wen-tau Yih scottyih@microsoft.com - Mohit Iyyer m.iyyer@gmail.com - Ming-Wei Chang minchang@microsoft.com ### 数据集概述 当前面向问答任务的语义解析研究多聚焦于冗长复杂的问题,其中多数问题在人类日常对话中显得不够自然。为探索对话式问答场景,我们提出了一项更贴近现实的任务:回答一系列简单但相互关联的问题。 我们通过让众包工作者拆解来自WikiTableQuestions(WTQ)的2022个问题构建了SQA数据集,该数据集包含针对维基百科表格的大量复合式问题。我们为每个WTQ问题安排三名众包工作者进行拆解,最终得到包含17553个问题的6066个问题序列。每个问题均关联表格中以单元格坐标形式标注的答案。 - Panupong Pasupat, Percy Liang. "半结构化表格上的复合语义解析" ACL-2015. [http://www-nlp.stanford.edu/software/sempre/wikitable/](http://www-nlp.stanford.edu/software/sempre/wikitable/) ### 支持任务与排行榜 [需要更多信息] ### 语言 英语(`en`)。 ## 数据集结构 ### 数据实例 {'id': 'nt-639', 'annotator': 0, 'position': 0, 'question': '这些球员来自哪里?', 'table_file': 'table_csv/203_149.csv', 'table_header': ['顺位', '球员', '球队', '位置', '毕业院校'], 'table_data': [['1', 'Ben McDonald', '巴尔的摩金莺队', '右投手', '路易斯安那州立大学'], ['2', 'Tyler Houston', '亚特兰大勇士队', '捕手', '"Valley HS (Las Vegas, NV)"'], ['3', 'Roger Salkeld', '西雅图水手队', '右投手', 'Saugus (CA) HS'], ['4', 'Jeff Jackson', '费城费城人队', '外野手', '"Simeon HS (Chicago, IL)"'], ['5', 'Donald Harris', '德州游骑兵队', '外野手', '德州理工大学'], ['6', 'Paul Coleman', '圣路易斯红雀队', '外野手', 'Frankston (TX) HS'], ['7', 'Frank Thomas', '芝加哥白袜队', '一垒手', '奥本大学'], ['8', 'Earl Cunningham', '芝加哥小熊队', '外野手', 'Lancaster (SC) HS'], ['9', 'Kyle Abbott', '加州天使队', '左投手', '长滩州立大学'], ['10', 'Charles Johnson', '蒙特利尔博览会队', '捕手', '"Westwood HS (Fort Pierce, FL)"'], ['11', 'Calvin Murray', '克利夫兰印第安人队', '三垒手', '"W.T. White High School (Dallas, TX)"'], ['12', 'Jeff Juden', '休斯顿太空人队', '右投手', 'Salem (MA) HS'], ['13', 'Brent Mayne', '堪萨斯皇家队', '捕手', 'Cal State Fullerton'], ['14', 'Steve Hosey', '旧金山巨人队', '外野手', '弗雷斯诺州立大学'], ['15', 'Kiki Jones', '洛杉矶道奇队', '右投手', '"Hillsborough HS (Tampa, FL)"'], ['16', 'Greg Blosser', '波士顿红袜队', '外野手', 'Sarasota (FL) HS'], ['17', 'Cal Eldred', '密尔沃基酿酒人队', '右投手', '爱荷华大学'], ['18', 'Willie Greene', '匹兹堡海盗队', '游击手', '"Jones County HS (Gray, GA)"'], ['19', 'Eddie Zosky', '多伦多蓝鸟队', '游击手', '弗雷斯诺州立大学'], ['20', 'Scott Bryant', '辛辛那提红人队', '外野手', '德克萨斯大学'], ['21', 'Greg Gohr', '底特律老虎队', '右投手', '圣克拉拉大学'], ['22', 'Tom Goodwin', '洛杉矶道奇队', '外野手', '弗雷斯诺州立大学'], ['23', 'Mo Vaughn', '波士顿红袜队', '一垒手', '塞顿霍尔大学'], ['24', 'Alan Zinter', '纽约大都会队', '捕手', '亚利桑那大学'], ['25', 'Chuck Knoblauch', '明尼苏达双城队', '二垒手', '德州农工大学'], ['26', 'Scott Burrell', '西雅图水手队', '右投手', 'Hamden (CT) HS']], 'answer_coordinates': {'row_index': [0, 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], 'column_index': [4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4]}, 'answer_text': ['路易斯安那州立大学', '"Valley HS (Las Vegas, NV)"', 'Saugus (CA) HS', '"Simeon HS (Chicago, IL)"', '德州理工大学', 'Frankston (TX) HS', '奥本大学', 'Lancaster (SC) HS', '长滩州立大学', '"Westwood HS (Fort Pierce, FL)"', '"W.T. White High School (Dallas, TX)"', 'Salem (MA) HS', 'Cal State Fullerton', '弗雷斯诺州立大学', '"Hillsborough HS (Tampa, FL)"', 'Sarasota (FL) HS', '爱荷华大学', '"Jones County HS (Gray, GA)"', '弗雷斯诺州立大学', '德克萨斯大学', '圣克拉拉大学', '弗雷斯诺州立大学', '塞顿霍尔大学', '亚利桑那大学', '德州农工大学', 'Hamden (CT) HS']} ### 数据字段 - `id`(字符串):问题序列ID(与WTQ中的ID保持一致) - `annotator`(整数):取值为`0`、`1`、`2`,代表标注该问题意图的三名标注者 - `position`(整数):问题在序列中的位置索引 - `question`(字符串):标注者提出的问题 - `table_file`(字符串):关联的表格文件路径 - `table_header`(字符串列表):表格的表头列表 - `table_data`(二维字符串数组):表格的实际数据内容 - `answer_coordinates`(字典列表):答案所在的表格单元格坐标(采用0索引,表头后的第一行索引为0) - `row_index`:行索引 - `column_index`:列索引 - `answer_text`(字符串列表):答案单元格的文本内容 注意:部分文本字段可能包含制表符或换行符,因此会以引号包裹。建议使用Python的CSV库等工具处理该数据集。 ### 数据划分 | | 训练集 | 测试集 | |-------------|-------:|-------:| | 样本数量 | 14541 | 3012 | ## 数据集构建 ### 构建初衷 [需要更多信息] ### 源数据 #### 初始数据收集与标准化 [需要更多信息] #### 源语言生产者是谁? [需要更多信息] ### 标注 #### 标注流程 [需要更多信息] #### 标注者是谁? [需要更多信息] ### 个人与敏感信息 [需要更多信息] ## 数据使用注意事项 ### 数据集的社会影响 [需要更多信息] ### 偏差讨论 [需要更多信息] ### 其他已知局限性 [需要更多信息] ## 附加信息 ### 数据集维护者 [需要更多信息] ### 许可信息 [微软研究数据许可协议](https://msropendata-web-api.azurewebsites.net/licenses/2f933be3-284d-500b-7ea3-2aa2fd0f1bb2/view)。 ### 引用信息 @inproceedings{iyyer-etal-2017-search, title = "基于搜索的神经结构化学习用于序列问答", author = "Iyyer, Mohit and Yih, Wen-tau and Chang, Ming-Wei", booktitle = "第55届计算语言学协会年会论文集(长文卷)", month = jul, year = "2017", address = "加拿大温哥华", publisher = "计算语言学协会", url = "https://aclanthology.org/P17-1167", doi = "10.18653/v1/P17-1167", pages = "1821--1831", } ### 贡献 感谢[@mattbui](https://github.com/mattbui)添加本数据集。
数据集概述
数据集名称
- 名称:Microsoft Research Sequential Question Answering (SQA)
- 别名:SQA
数据集基本信息
- 语言:英语 (
en) - 许可证:Microsoft Public License (MS-PL)
- 多语言性:单语种
- 大小:10K<n<100K
- 来源:原创数据集
- 任务类别:问答
- 任务ID:抽取式问答 (
extractive-qa)
数据集内容
- 创建目的:探索对话式问答设置,处理一系列简单但相互关联的问题。
- 数据来源:从WikiTableQuestions(WTQ)中分解的2,022个问题,由三名工作人员分解,形成6,066个序列,包含17,553个问题。
数据集结构
-
特征:
id(字符串):问题序列IDannotator(整数):标注者IDposition(整数):问题在序列中的位置question(字符串):问题内容table_file(字符串):关联的表格文件table_header(字符串序列):表格头部信息table_data(字符串序列的序列):表格数据answer_coordinates(序列):答案在表格中的坐标row_index(整数)column_index(整数)
answer_text(字符串序列):答案内容
-
数据分割:
- 训练集:12276个例子,19732499字节
- 验证集:2265个例子,3738331字节
- 测试集:3012个例子,5105873字节
数据集创建
-
许可证信息:Microsoft Research Data License Agreement
-
引用信息:
@inproceedings{iyyer-etal-2017-search, title = "Search-based Neural Structured Learning for Sequential Question Answering", author = "Iyyer, Mohit and Yih, Wen-tau and Chang, Ming-Wei", booktitle = "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = jul, year = "2017", address = "Vancouver, Canada", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/P17-1167", doi = "10.18653/v1/P17-1167", pages = "1821--1831", }
数据集使用注意事项
- 数据处理:建议使用CSV解析器(如Python的CSV包)处理数据,因为某些文本字段可能包含Tab或LF字符。




