Similar data elimination: MFB algorithm

Nowadays, the complex applications such as knowledge extraction, data mining, E-learning and web applications use heterogeneous and distributed data. In this context, the need for integration and improving data quality is increasingly felt. The problem of eliminating duplicates and similar data is still relevant in terms of both performance and in terms of the definition of similarity rules. We present in this paper a new deduplication algorithm based on the two functions Match and Merge. An evaluation is made experimentally using a set of randomly generated data.