This article describes a Data-Mining-based application intended for the financial sector. Such an application evaluates the level of risk associated to financial loans on approval. A sample of 1000 data records from a commercial bank were analyzed and further processed. Knowledge Discovery in Databases (KDD) methodology was implemented and a software tool that allows discrete-data conversion was developed so that the samples could be used as input data to the data mining tool called WEKA. Results were compared to assess the performance when applying data mining techniques and classification trees ID3 and J48. Finally, the application yields the characteristics that customers should exhibit to be granted financial loans.