@inproceedings{5ffb9d8c8cf9413db39b8dec090be34d,
title = "Efficient mapReduce-based method for massive entity matching",
abstract = "Most of the state-of-the-art MapReduce-based entity matching methods inherit traditional Entity Resolution techniques on centralized system and focus on data blocking strategies in order to solve the load balancing problem occurred in distributed environment. In this paper, we propose a MapReduce-based entity matching framework for processing semi-structured and unstructured data. We use a Locality Sensitive Hash (LSH) function to generate low dimensional signatures for high dimensional entities; we introduce a series of random algorithms to ensure that similar signatures will be matched in reduce phase with high probability. Moreover, our framework contains a solution for reducing redundant similarity computation. Experiments show that our approach has a huge advantage on processing speed whilst keeps a high accuracy.",
author = "Pingfu Chao and Zhu Gao and Yuming Li and Junhua Fang and Rong Zhang and Aoying Zhou",
note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 16th International Conference on Web-Age Information Management, WAIM 2015 ; Conference date: 08-06-2015 Through 10-06-2015",
year = "2015",
doi = "10.1007/978-3-319-21042-1\_48",
language = "英语",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "494--497",
editor = "Jian Li and Yizhou Sun",
booktitle = "Web-Age Information Management - 16th International Conference, WAIM 2015, Proceedings",
address = "德国",
}