Generating and applying a prediction model based on blockchain data

A method, comprising:

• obtaining, from at least a first node of a plurality of nodes, a first set of information associated with a first plurality of invoices;
• wherein each of the plurality of nodes maintains a respective copy of a same blockchain ledger;
• wherein the blockchain ledger comprises a plurality of blocks, each block comprising one or more transactions, and each block associated with a preceding block based on a hash of the preceding block;
• wherein the blockchain ledger comprises at least one transaction associated with each of the first plurality of invoices;
• iteratively training a machine learning algorithm using the first set of information associated with the first plurality of invoices as training data, wherein the machine learning algorithm generates a prediction model;
• determining, using the prediction model, a predicted lowest cost of obtaining funding using an invoice and an associated funding type;
• obtaining a target funding amount and a target funding date of a particular entity;
• generating a recommendation for obtaining the target funding amount, by the target funding date, using a subset of a second plurality of invoices issued by the particular entity, wherein the generating the recommendation comprises:
• obtaining, from at least a second node of the plurality of nodes, a second set of information associated with a first invoice of the second plurality of invoices;
• obtaining, from at least a third node of the plurality of nodes, a third set of information associated with a second invoice of the second plurality of invoices;
• determining, by applying the prediction model to the second set of information (a) a first predicted lowest cost of obtaining funding, by the target funding date, using the first invoice and (b) a first associated funding type;
• determining, by applying the prediction model to the third set of information (a) a second predicted lowest cost of obtaining funding, by the target funding date, using the second invoice and (b) a second associated funding type;
• wherein the first associated funding type and the second associated funding type are different;
• based on a comparison of the first predicted lowest cost and the second predicted lowest cost, selecting the first invoice for obtaining the target funding amount of the particular entity, without selecting the second invoice for obtaining the target funding amount of the particular entity;
• and
• presenting, at an interface, the recommendation for obtaining the target funding amount by the target funding date using the first invoice, without using the second invoice;
• wherein the method is performed by at least one device including a hardware processor.

A system, comprising:

• at least one device including a hardware processor;
• and
• the system performing operations comprising:
• obtaining, from at least a first node of a plurality of nodes, a first set of information associated with a first plurality of invoices;
• wherein each of the plurality of nodes maintains a respective copy of a same blockchain ledger;
• wherein the blockchain ledger comprises a plurality of blocks, each block comprising one or more transactions, and each block associated with a preceding block based on a hash of the preceding block;
• wherein the blockchain ledger comprises at least one transaction associated with each of the first plurality of invoices;
• iteratively training a machine learning algorithm using the first set of information associated with the first plurality of invoices as training data, wherein the machine learning algorithm generates a prediction model;
• determining, using the prediction model, a predicted lowest cost of obtaining funding using an invoice and an associated funding type;
• obtaining a target funding amount and a target funding date of a particular entity;
• generating a recommendation for obtaining the target funding amount, by the target funding date, using a subset of a second plurality of invoices issued by the particular entity, wherein the generating the recommendation comprises:
• obtaining, from at least a second node of the plurality of nodes, a second set of information associated with a first invoice of the second plurality of invoices;
• obtaining, from at least a third node of the plurality of nodes, a third set of information associated with a second invoice of the second plurality of invoices;
• determining, by applying the prediction model to the second set of information to determine (a) a first predicted lowest cost of obtaining funding, by the target funding date, using the first invoice and (b) a first associated funding type;
• determining, by applying the prediction model to the third set of information to determine (a) a second predicted lowest cost of obtaining funding, by the target funding date, using the second invoice and (b) a second associated funding type;
• wherein the first associated funding type and the second associated funding type are different;
• based on a comparison of the first predicted lowest cost and the second lowest predicted cost, selecting the first invoice for obtaining the target funding amount of the particular entity, without selecting the second invoice for obtaining the target funding amount of the particular entity;
• and
• presenting, at an interface, the recommendation for obtaining the target funding amount by the target funding date using the first invoice, without using the second invoice.