We propose a deep Siamese neural network that learns a similarity measure tailored to a dataset, eliminating manual feature engineering. We also show that knowledge transfer …
We propose the first workflow that systematically integrates data preparation operations before duplicate detection, improving AUC-PR by up to 19%.
In this paper, we study the problem of matching records that contain address information, including attributes such as Street-address and City. To facilitate this matching process …
We propose an ensemble-based partitioning method to improve theta-join execution in massively parallel systems such as MapReduce and Spark.
Collaborative recommenders rely on the assumption that similar users may exhibit similar tastes while content-based ones favour items that found to be similar with the items a user …