Coupa Software Incorporated

In an embodiment, a computer-implemented method comprises storing invoice data comprising a plurality of invoice objects, each invoice object comprising a digital programmatic object that is maintained in computer memory, each invoice object comprising one or more data fields that are programmatically linked to and storing data for an invoice between a supplier account and a buyer account, the one or more data fields including a remittance address that specifies an address to which a payment is to be sent; determining, based on the plurality of invoice objects, a set of unique remittance addresses, each unique remittance address of the set of unique remittance addresses associated with a distinct supplier account; generating a first set of payment objects based on the set of unique remittance addresses, each payment object of the first set of payment objects associated with a unique remittance address of the set of unique remittance addresses; determining, based on the plurality of invoice objects, that a particular unique remittance address of the set of unique remittance addresses is associated with multiple buyer entities of a particular buyer account; in response to determining that the particular unique remittance address of the set of unique remittance addresses is associated with the multiple buyer entities of the particular buyer account, generating a second set of payment objects, each payment object of the second set of payment objects associated with the particular unique remittance address and a distinct buyer entity of the multiple buyer entities; causing multiple payment transactions based on the first set of payment objects and the second set of payment objects.

In an embodiment, a method for dynamically generating and using peer groups in an e-procurement system includes creating an entity database with a plurality of attributes associated with entities and an ordered table of matching rules. Each matching rule having a priority value and two or more matching attributes. The method includes receiving a user input specifying a particular entity for generating a current peer group of other entities. The method includes accessing the entity database to retrieve particular attributes of the particular entity and querying the entity database to receive a result set of matching entities as the current peer group of other entities for the particular entity based on the particular attributes of the particular entity and the matching rules. The method further includes appending data for the matching entities in the result set to a peer group data structure associated with the particular entity.

Input data comprises an order to provide a quantity of one or more finished goods to a site. Software is programmed for: accessing information that defines an architecture of a supply chain network comprising a plurality of sites connected by a plurality of path fragments; generating one or more path solutions to fulfill the order; removing one or more loops from the one or more path solutions; generating based on the one or more path solutions and the one or more removed loops, one or more revised path solutions; outputting the one or more revised path solutions for display on a client device; and transmitting, only to computers of the sites of a revised subset of a plurality of path fragments, a plurality of signals to initiate movement of one or more of raw materials or finished goods.

Systems and methods for presenting configurable machine learning systems through graphical user interfaces are disclosed. In an embodiment, a machine learning server computer stores one or more machine learning configuration files. A particular machine learning configuration file of the one or more machine learning configuration files comprises instructions for configuring a machine learning system of a particular machine learning type with one or more first machine learning parameters. The machine learning server computer displays through a graphical user interface, a plurality of selectable parameter options, each of which defining a value for a machine learning parameter. The machine learning server computer receives a particular input dataset. The machine learning server computer additionally receives, through the graphical user interface, a selection of one or more selectable parameter options corresponding to one or more second machine learning parameters different from the one or more first machine learning parameters. The machine learning server computer replaces in the particular machine learning configuration file, the one or more first machine learning parameters with the one or more second machine learning parameters. Using the particular machine learning configuration file, the machine learning server computer configures a particular machine learning system. Using the particular machine learning system and the particular input dataset, the machine learning server computer computes a particular output dataset.

Embodiments disclosed herein may provide capabilities for multi-source data gathering, adaptive item cross referencing, data preparation and data extraction. These capabilities may allow the creation item master records which can provide richer information than available from any one discrete source. Additional functionality which may be provided in some embodiments could also include providing commodity based predictive pricing and/or a visual spend map.

In an embodiment, a computer-implemented method comprises receiving supplier invoice data comprising one or more supplier payment objects corresponding to a supplier to a customer of a supplier-customer transaction. A match determination is made between the supplier payment object and a verified payment object. In response to a no-match determination, a customer transaction system is queried to validate the supplier, and in response to the supplier validation, a validation payment object is created, validating the supplier. A match determination is made between the supplier payment object and the verified payment object, and the supplier invoice is automatically generated. In response to receiving subsequent supplier invoice data corresponding to the supplier of subsequent supplier-client transactions, for each subsequent supplier-client transaction, the subsequent supplier invoice is automatically generated with the verified payment object.

In one embodiment, a computer-implemented method comprises, using an e-procurement platform instance that is programmed to host thousands of user accounts that are associated with a plurality of different entities, receiving, from a buyer computer, supplier payment account data for a dataset representing a payment to a supplier; creating a cryptographically secure fingerprint based on the supplier payment account data; using the fingerprint and at least a portion of the supplier payment account data, programmatically transmitting a structured database query to a digital data repository, the digital data repository comprising at least hundreds of thousands of records of transactions that included other datasets representing other payments; in response to the querying, receiving a result set of a plurality of supplier payment datasets, based on the records of transactions, the supplier payment datasets representing past submissions of payments to the same supplier payment account that the fingerprint represents; based on the result set and a plurality of programmatic rules, calculating one or more notification metrics and selecting a notification from among a plurality of programmed notifications and transmitting, to the buyer computer, presentation instructions which when rendered using the buyer computer cause displaying the one or more notification metrics and a notification text.

A computer-implemented method for detecting and managing duplicate invoices is provided. In one embodiment, the method includes accessing, in a digital storage device, invoices associated with a buyer computer. Candidate invoices having similar attributes are identified from the invoices and nodes representing the candidate invoices are created and stored in computer memory. At least two of the candidate invoices are determined likely to be duplicate invoices and an edge is generated between nodes representing the at least two candidate invoices, indicating that the at least two candidate invoices are likely to be duplicate invoices. A set of filters is programmatically applied using the at least two candidate invoices as inputs. The at least two candidate invoices are determined to be duplicate invoices based on an output of the set of filters and a notification is sent to the buyer computer indicating that the at least two candidate invoices are duplicate invoices.

Computer-implemented techniques for repeatable and interpretable divisive analysis. In one embodiment, for example, a method comprises: identifying top-level cohorts of data items based on one or more characteristics of the data items in common; recursively or iteratively dividing a selected top-level cohort in a top-down manner resulting in a plurality of sub-level cohorts arranged in a hierarchy; detecting a particular data item that is a statistical outlier among data items of a leaf cohort in the hierarchy; and causing display of an indication in a computer user interface that the particular data item is an outlier.

In one embodiment, a method includes assessing unlabeled transaction data and unlabeled candidate transaction data. The method includes determining a first plurality of features for the unlabeled transaction data and a second plurality of features for the unlabeled candidate transaction data using a Bag of Words model. The method includes determining matching pairs using the unlabeled transaction data and the unlabeled candidate transaction data. The method includes calculating similarity scores for the matching pairs using the first plurality of features and the second plurality of features. The method includes generating a machine learning model to predict a subcategory label and a ranking value for the unlabeled candidate transaction data using the similarity scores of the matching pairs. The method includes determining, using the machine learning model, the subcategory label and the ranking value associated with the unlabeled candidate transaction data.