Abstract: Deterministic and probabilistic are two approaches to matching commonly used in Entity Resolution (ER) systems. While many users are familiar with writing and using Boolean rules for deterministic matching, fewer are as familiar with the scoring rule configuration used to support probabilistic matching. This paper describes a method using deterministic matching to “bootstrap” probabilistic matching. It also examines the effectiveness three commonly used strategies to mitigate the effect of missing values when using probabilistic matching. The results based on experiment using different sets of synthetically generated data processed using the OYSTER open source entity resolution system.

Awaad Alsarkhi and John R. Talburt, “A Method for Implementing Probabilistic Entity Resolution” International Journal of Advanced Computer Science and Applications(IJACSA), 9(11), 2018. http://dx.doi.org/10.14569/IJACSA.2018.091102

@article{Alsarkhi2018,
title = {A Method for Implementing Probabilistic Entity Resolution},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2018.091102},
url = {http://dx.doi.org/10.14569/IJACSA.2018.091102},
year = {2018},
publisher = {The Science and Information Organization},
volume = {9},
number = {11},
author = {Awaad Alsarkhi and John R. Talburt}
}