Abstract
Neural networks have been successfully employed in a wide variety of fields, such as signal processing, pattern recognition, medicine, speech recognition, and business, in order to solve complex problems. It is proposed in this paper that neural networks can also be applied to the data quality problem that is so pervasive in legacy software systems. The focus of this work is on the use- of neural networks to learn how to identify duplicate records in a data source. This problem has been recognized as extremely important to many organizations, due to the size and complexity of today database systems. The initial use of neural nets has shown that they can be trained to perform data quality checks. Our ongoing research will address larger, more complex databases systems, as well as, learning capabilities to solve other data quality problems.