Lendingclub Bank, National Association

Jianglan Han

Jianju Liu

Wenxuan Xu

Aaron Jimenez

Dhwani Umeshbhai Bosamiya

Phani Pradeep Benarji Kommana

Shi Tang

Techniques are provided for forming clusters of individual prediction targets (IPTs). An initial prediction target is a target for which an automated prediction has been generated. IPTs may be, for example, borrowers to which a lending entity has extended loans based on predictions generated by a credit policy. Each cluster includes (a) a 'core' of underperforming entities (UEs), and (b) a set of boundary performant entities (PEs). The UEs that belong to the UE core of a cluster are “similarly situated” relative to the values of their features. For example, in the context where the IPTs are borrowers, the UEs at the core of a cluster may correspond to defaulting borrowers that had similar bureau data, lending entity data, and borrower data. The boundary performant entities of the cluster may be borrowers that have not defaulted, but had similar credit qualifications as the UEs of the cluster. Having formed these clusters, the clusters may be used in a variety of ways, including but not limited to improving the accuracy of the credit model, identifying potentially problematic future borrowers, generating visualizations that illustrate the relative importance of clusters of defaulting borrowers, etc.

Techniques for improving the accuracy of automated predictions

Techniques are provided for tracking a number of transactions-of-interest in a transaction-recording blockchain using a control blockchain. A transaction-of-interest is a transactions that is in a particular state. A request to perform an action is received. Upon receiving the request and determining that the action corresponds to a new transaction-of-interest, a control blockchain is checked to determine the current number of transactions-of-interest in the transaction-recording blockchain and maximum allowed number of transaction-of-interest for the transaction-recording blockchain. In response to determining that the current number of transaction-of-interest in the transaction-recording blockchain are less than the maximum allowed: allowing the action to occur, adding a new block to the transaction-recording blockchain, and updating the control blockchain to indicate the new number of transaction-of-interest. In response to determining that the current number of transaction-of-interest are at least at the maximum allowed number, prohibiting the addition of a new block to the transaction-recording blockchain.

Restricted blockchain cluster

Growing community of inventors

San Francisco, CA, United States of America

Jianglan Han