Leopoldo Bertossi
Carleton University. Ottawa Canada
email: bertossi@scs.carleton.ca
Carleton University. Ottawa Canada
In this presentation we give an introduction to the combination of matching dependencies for entity resolution and support-vector machines, a classification method in machine learning. We show that they can be used to address the problems of identifying duplicate representations in data sources and merging them into single representations. The logical glue is provided by Datalog, a logical query language for relational databases, in its incarnation as LogiQL, which has been developed by LogicBlox. This presentation emphasizes the current trend of logically combining different techniques for addressing different interrelated problems in data management.
The presentation will be given in Spanish.
Short Biography Leopoldo Bertossi has been Full Professor at the School of Computer Science, Carleton University (Ottawa, Canada) since 2001. He is also a Faculty Fellow of the IBM Center for Advanced Studies (IBM Toronto Lab). He has been a professor at the Department of Computer Science, PUC-Chile (until 2001); and also the President of the Chilean Computer Science Society (SCCC). His research interests include data management in general, database theory, business intelligence, data quality, semantic web data, logic-based knowledge representation, and machine learning. He obtained a PhD in Mathematics from the PUC-Chile in 1988.
@InProceedings{CLEI-2015:KN-Bertossi,
author = {Leopoldo Bertossi},
title = {Combining Matching Dependencies and Machine Learning via Datalog for Entity Resolution in Databases},
booktitle = {2015 XLI Latin American Computing Conference (CLEI), Special Edition},
pages = {4--4},
year = {2015},
editor = {Universidad Católica San Pablo},
address = {Arequipa-Peru},
month = {October},
organization = {CLEI},
publisher = {CLEI},
url = {http://clei.org/clei2015/KN-Bertossi},
isbn = {978-9972-825-91-0},
}