Abstract: A description of a machine learning (ML) model is received, with the ML model including multiple features such as an unlikely combination feature, which corresponds to a first attribute to be located in an invoice and a second attribute to be located the invoice concurrently with the first attribute. Training data is received, including (i) invoice data with multiple invoices, each including the first attribute and the second attribute, and respective values of the first attribute and the second attribute, and (ii) validity data including indications of which of the invoices are valid and which of the invoices are invalid. The ML model is trained using the training data using the ML model. The training includes applying the values of the attributes to the unlikely combination feature. The ML model is applied to an invoice to be validated to determine a probability that the invoice is invalid.

Abstract: Systems and methods are provided for manipulating objects in a framework software application that embeds another software application that does not natively support object manipulation controls of the framework software application. To overcome this difficulty, a user interface of the embedded software application is provided in an embedded window disposed within a framework window. Moreover, the user interface of the framework software application is provided in the framework window. Next, a transparent interface element, configured to detect events generated by the object manipulation controls of the framework software application, is generated, and is positioned over the embedded window.

Abstract: A description of a machine learning (ML) model is received, with the ML model including multiple features such as an unlikely combination feature, which corresponds to a first attribute to be located in an invoice and a second attribute to be located the invoice concurrently with the first attribute. Training data is received, including (i) invoice data with multiple invoices, each including the first attribute and the second attribute, and respective values of the first attribute and the second attribute, and (ii) validity data including indications of which of the invoices are valid and which of the invoices are invalid. The ML model is trained using the training data using the ML model. The training includes applying the values of the attributes to the unlikely combination feature. The ML model is applied to an invoice to be validated to determine a probability that the invoice is invalid.