CrowdSR: A Crowd Enabled System for Semantic Recovering of Web Tables

被引：5

作者：

Liu, Huaxi ^{[1
]}

Wang, Ning ^{[1
]}

Ren, Xiangran ^{[1
]}

机构：

[1] Beijing Jiaotong Univ, Sch Comp & Informat Technol, Beijing, Peoples R China

来源：

WEB-AGE INFORMATION MANAGEMENT (WAIM 2015) | 2015年 / 9098卷

关键词：

D O I：

10.1007/978-3-319-21042-1_67

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Without knowing any semantic of tables on web, it's very difficult for web search to take advantage of those high quality sources of relational information. We present CrowdSR, a system that enables semantic recovering of web tables by crowdsourcing. To minimize the number of tuples posed to the crowd, CrowdSR selects a small number of representative tuples by clustering based on novel integrative distance. An evaluation mechanism is also implemented on Answer Credibility in order to recommend related tasks for workers and decide the final answers for each task more accurately.

引用

页码：581 / 583

页数：3