Methods, systems, and computer programs for matching a transaction against data in a database are provided to identify the transaction parameters based on text descriptors provided for the transaction. A method comprises a process for identifying features for matching transactions of a first functional unit by means of a machine learning program. The features include at least a description of the transaction, a name of a second functional unit in the transaction, a location of the second functional unit and an account for the transaction. The machine learning program is trained with training data that include values of the characteristics for previously reconciled transactions. A received first transaction includes a description, a date and an amount. The first transaction is entered for the machine learning program that generates one or more suggestions for matching the first transaction. Each proposal includes the name of the second functional unit in the first transaction and an account.
展开▼