Ιδρυματικό Αποθετήριο
Πολυτεχνείο Κρήτης
Πολυτεχνείο Κρήτης
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/284C74C9-4024-4E15-9E13-754BFF15CE0D | ||
| Έτος | 2009 | ||
| Τύπος | Πλήρης Δημοσίευση σε Συνέδριο | ||
| Άδεια Χρήσης |
|
||
| Βιβλιογραφική Αναφορά | V. Rastogi, N. Dalvi and M. Garofalakis, "Large-scale collective entity matching", in 35th International Conference on Very Large Data Bases, 2009. https://doi.org/10.14778/1938545.1938546 | ||
| Εμφανίζεται στις Συλλογές |
There have been several recent advancements in Machine Learningcommunity on the Entity Matching (EM) problem. However,their lack of scalability has prevented them from being applied inpractical settings on large real-life datasets. Towards this end, wepropose a principled framework to scale any generic EM algorithm.Our technique consists of running multiple instances of the EM algorithmon small neighborhoods of the data and passing messagesacross neighborhoods to construct a global solution. We prove formalproperties of our framework and experimentally demonstratethe effectiveness of our approach in scaling EM algorithms.
Copyright © DIAS 2013
