Bank of America Corporation

Yanghong Shao

David Joa

Thayer S Allison

Debashis Ghosh

Kurt Newman

Timothy J Bendel

Sudeshna Banerjee

Mark V Krein

Naveen Gururaja Yeri

David Neil Joffe

Arun Pinto

Deepak Verma

Jade Michelle Le Vo-Dinh

Bibhudatta Jena

Srihari Rao Gatpa

Preston Wesley Ports, Iii

Vikas Mishra

Erik Stephen Ross

Hans Dirk Schumacher

Timothy Lamonte Atwell

Marc Douglas Halsted

Sreedevi Gummuluri

Seyamak Amin

Elaine C Marino

Sriram Mohanraj

Edward W Carroll

Ryan M Greenway

Robert Caruso

Arya Kumar Vedabrata

Matthew Rice

Jangaiah Kranti Kurva

Joy M Tarquin

Chandrashekar Katuri

Maryann Mangini

Moncef El Ouriaghli

Nishitha Kakani

Parul Bajaj

Kunal Makin

Dilip Nair

Tracy Elizabeth Marcin

John Scott Bailer

Artificial-intelligence computer-implemented processes and machines predict whether personal data may be present in structured software based on metadata field(s) contained therein. Natural language processing preprocesses input strings corresponding to the metadata field(s) into normalized input sequence(s). Individual characters in the sequence(s) are embedded into fixed-dimension vectors of real numbers. Bidirectional LSTM(s) or other machine-learning algorithm(s) are utilized to generate forward and backward contextualization(s). Neural network output(s) are provided based on element-wise averaging or feed forwarding based on the contextualization(s) in order to predict whether one or more value fields corresponding to the metadata field(s) may contain personal data.

Personal data discovery

A computing platform may generate requests for log data stored on computing systems and may communicate the requests to the computing systems. The computing platform may receive the log data from the computing systems and may utilize the log data to generate records interrelating different data-access requests indicated by the log data. The computing platform may analyze the records to identify corresponding requests made by a user to multiple different computing systems and may determine an interrelationship between the corresponding requests. The computing platform may generate data indicating the multiple different computing systems and the interrelationship between the corresponding requests and may communicate the data to a computing device associated with at least one of the multiple different computing systems.

Inter-system data forensics

Systems, methods and computer program products are provided for optimizing social networks. In the systems and methods, account data associated with a user account is received and the account data is stored in a storage device. Pluralities of networking opportunities are identified based on the account data. Furthermore, based on the account data triggers may be detected indicating a user's interest in one or more social networking opportunities. The system then offers the user, based at least partially on the detected trigger, the one or more social networking opportunities.

System for optimizing social networking

Systems and methods for monitoring trigger data quality are provided herein. The systems and methods detect and report whether the current trigger counts are normal or flawed in real time. The systems and methods monitor the triggers to determine the accuracy, completeness, domain of values, and format of the trigger data. In the system and methods, account data associated with the one or more accounts is received and stored in a storage device; the account data is segregated into one or more periods of time; triggers associated with the one or more periods of time are identified based on transactions that occur during the one or more periods of time; a total transaction count for each of the triggers is calculated; and control limits are determined based on the transaction count for each of the triggers.

Trigger data quality monitor

Aspects of this disclosure relate to a computer configured to identify potential customers which may include a processor and memory storing computer executable instructions that, when executed, cause the computer to perform a method for identifying potential customers for a first bank, by electronically receiving data regarding the first bank, electronically receiving data regarding a second bank. Further, based on the electronically received data the method may further include identifying locations of the first bank, identifying locations of the second bank, comparing the locations of the first bank with the locations of the second bank, determining locations of the second bank that are within a predetermined distance of at least one location of the first bank, determining an amount of funds in a set of deposit accounts in each location of the second bank, and determining if the amount of funds in each location of the second bank is more than a predetermined amount.

Identifying potential customers using payment information

Systems and methods for producing and maintaining account related triggers are provided herein. The systems and methods may be utilized for risk mitigation. Account triggers may be executed to determine account activity indicative of a risk event. Once a risk event is identified, the financial institution may determine an appropriate action to mitigate foreseeable risk due to the risk event.

Financial account related trigger feature for risk mitigation

Apparatus and methods for processing a loan application. The apparatus and methods may involve accessing a financial institution customer account that includes a plurality of transactions; identifying those transactions that belong to a predetermined transaction category; and summarizing the transactions. Summary information may be presented as part of an income and expenses flow report that at least partially characterizes the account or the applicant. A financial institution representative may query a database for the summary information and, optionally, initiate an underwriting process based on the loan application and the summary information.

Apparatus and methods for verifying a credit applicant's income that enhance a credit applicant's experience

Systems, methods, and computer program products are provided for optimizing amounts collected in an offset action where a creditor attaches funds from a debtor's deposits account to satisfy overdue debt payments. In operation, embodiments of the present invention leverage a deposits account's transaction history to detect a deposit pattern for the account and, based on the detected pattern, predict the date of the next large deposit to the attachable account. An offset associate, using the present invention, can then schedule the offset action to occur on the expected date of the next large deposit.

Offset optimization system

Systems and methods for producing and maintaining account related triggers are provided herein. The systems and methods may be utilized for monitoring of one or more accounts of a financial institution. Triggers may be executed to monitor financial account ratios. Offers may be triggered based on the account ratios or changes in the account ratios.

Financial account related trigger feature for detecting changes in financial ratios

Aspects of the disclosure relate to assisting credit account holders during economic hardship. More specifically, aspects of the disclosure relate to systems and methods for providing a rebate of at least a portion of a finance charge applied to a credit account. Unlike certain prior art systems and methods, embodiments disclosed herein may benefit both the credit account holder as well as the financial institution that extends the credit account. In one embodiment, it may be determined that several credit accounts meet a threshold of probability of having a charge-off. In certain embodiments, the number of accounts that receive a rebate may be limited to a certain percentage of credit accounts with a non-zero balance.

Financial charge rebate

In general, embodiments of the invention relate to systems, methods, and computer program products for predicting population migration and analyzing migration-affecting programs. For example, an apparatus is provided having a memory device with population information and a plurality of migration factors stored therein. The population information includes information about population distribution across a plurality of classifications. Each of the plurality of migration factors corresponds to a particular classification of the plurality of classifications and indicates how population members of the particular classification migrate to other classifications over a particular time period. The apparatus also includes a processor communicably coupled to the memory device and configured to use the migration factors and the population information to forecast changes in the population distribution across each of the plurality of classifications over multiple time periods. In one embodiment, the systems, methods, and computer program products are configured to predict lateness characteristics of a lending portfolio.

Lateness migration model

Behavior associated with credit misuse is referred to herein as 'intentional misuse of credit' (IMOC) behavior and individuals who exhibit IMOC behavior are referred to herein as IMOC individuals. Embodiments of the present invention provide systems, methods, and computer program products for reviewing, on a daily basis, financial-transaction data taken across a plurality of individual consumers, identifying individual consumers whose financial-transaction data indicates IMOC behavior, and flagging those individual consumers as IMOC individuals. To enhance financial institutions' ability to make accurate credit-approval decisions, other embodiments of the present invention provide systems, methods, and computer program products for reviewing a particular credit applicant's most recent financial-transaction data and, based thereon, determining whether that credit applicant, who otherwise has a good credit history, exhibits recent IMOC behavior and thereby possesses the propensity to misuse the applied-for credit.

Detecting credit misuse

Techniques for bridging data from business support units, e.g., call centers and marketing operations, with a data warehouse to augment and enrich pre-existing customer information. Unstructured data is received with incomplete integration information. A data key is created from the unstructured data, and the unstructured data is integrated with the structured data in a data warehouse based on the data key. Unstructured data can assume different forms of data, including recorded audio data, facial image data, and iris image data. At least one customer identifier is extracted from the unstructured data by data mining. A data key is subsequently created from the at least one customer identifier. Incomplete integration may include partial integration information or may not include any integration information.

Unstructured data integration with a data warehouse

Systems and methods that may determine whether to provide a resource to a consumer are disclosed. Further methods may allow distributing resources on an expedited basis or recheck previous determinations to provide a resource. In certain embodiment, resource transactions may be categorized. Illustrative categories may be organized by: investment, a non-investment expenditure, or income. In one embodiment, transactions categorized as non-investment expenditures may be categorized into subcategories, including necessities, discretionary, or debt. Calculations may be performed upon the categorized (and/or sub-categorized data). For example, one or more of the following ratios may be calculated: debt-to-income, investment-to-income, and expenditure-to-income. Ratios may be compared to one or more threshold values. Determinations of the threshold may consider information regarding other individuals' or business' resource transactions.

Determinations relating to resource distribution

Systems, methods, and computer program products are provided for optimizing amounts collected in an offset action where a creditor garnishes funds from a debtor's deposits account to satisfy delinquent debt payments. In operation, embodiments of the present invention leverage a deposits account's transaction history to detect a deposit pattern for the account and, based on the detected pattern, predict the date of the next large deposit to the garnishable account. An offset associate, using the present invention, can then schedule the offset action to occur on the expected date of the next large deposit.

Growing community of inventors

Charlotte, NC, United States of America

Yanghong Shao