Association Rule Mining from large datasets of clinical invoices document

The concept of massive data generation nowadays affects several domains such as marketing including electronic invoices of large retailers, web access log files, healthcare, life sciences and so on. All these web activities introduced a new way to pay through the concept of electronic invoices (eInvoice), replacing the paper invoices. For these reasons, eInvoicing can be thought of as an innovative digital infrastructure for the issue, transmission, and storage of invoices. The availability of large volumes of eInvoices allows the discovery of new knowledge through data mining in these domains. Thus, users by using data mining can extract knowledge from large invoices documents. In this paper, we present a software tool for mining association rules from invoices produced in healthcare centers. In particular, the tool adopt a novel preprocessing methodology that provides merging, cleaning, formatting and summarization of eInvocies. The methodology can improve the quality of a huge amount of clinical invoices reducing the quantity of irrelevant data, making the remaining data suitable to mine information in form of association rules. The core of the tool allows to extract association rules from eInvoices; as a case study, we discuss the mined rules, highlighting the relationships among the purchased goods.