US20210142401A1

US20210142401A1 - Customizing loan specifics on a per-user basis

Info

Publication number: US20210142401A1
Application number: US15/499,203
Authority: US
Inventors: Elizabeth Flowers; Puneit Dua; Alan Zwilling; Adam Mattingly; Melissa Attig; Reena Batra
Original assignee: State Farm Mutual Automobile Insurance Co
Current assignee: State Farm Mutual Automobile Insurance Co
Priority date: 2016-05-05
Filing date: 2017-04-27
Publication date: 2021-05-13
Also published as: US20220027951A1; US20220138804A1; US11004116B1; US20240169399A1; US11257122B1; US10891655B1; US10891628B1; US20220083995A1; US20220051292A1; US20230206316A1; US20210150625A1; US20170323345A1; US10977725B1; US11900421B2; US10733631B2; US20210358030A1

Abstract

Techniques are disclosed to provide customized loans on a per-user basis. With user permission or affirmative consent, user data may be monitored for several users, which may be used to calculate initial loan specifics such as a loan rate and term based upon a portion of this user input data. The user data may include demographic data, behavioral data, or other data indicative of a user's future potential earnings or other relevant information that may be analyzed to determine, for that specific user, the current likelihood that the user will default on the loan and a future likelihood of default. When this future statistical likelihood is determined, the initial loan specific may be further modified and/or a targeted notification may be sent indicating these customized loan specifics.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to (1) Provisional Application No. 62/338,749, entitled “Using Cognitive Computing To Customize Loans,” filed on May 19, 2016; (2) Provisional Application No. 62/332,226, entitled “Using Cognitive Computing To Provide a Personalized Banking Experience,” filed on May 5, 2016; (3) Provisional Application No. 62/338,752, entitled “Using Cognitive Computing To Provide a Personalized Banking Experience,” filed on May 19, 2016; and (4) Provisional Application No. 62/341,677, entitled “Using Cognitive Computing To Improve Relationship Pricing,” filed on May 26, 2016, each of which is incorporated herein by reference in its entirety.

TECHNICAL FIELD

The present disclosure relates to providing an improved banking experience and, more particularly, to customizing the details and specifics of loans on a per-user basis.

BACKGROUND

Traditionally, to apply for various types of loans, people would visit a brick-and-mortar bank or other financial institution to meet with a loan officer or other representative, submit the necessary documentation and, if approved, sign a loan agreement defining the loan specifics, such as the loan term and rate. More recently, the loan application and approval process has been simplified by providing users with online options to apply, and potentially be approved for, various types of loans. Therefore, because of the ease in which people may apply for loans and compare interest rates among different lenders, the competition amongst lenders has dramatically increased.
This competition has increased the importance of advertising to users and identifying potential clients that may require loans, but the personalization of these loans leaves much to be desired. For instance, although the details of a loan such as the term and interest rate may be calculated using a user's credit score, this calculation does not take into account any potential future impact on this credit score or finer details that could otherwise impact these loan specifics. Therefore, traditional techniques directed to calculating loan specifics for a user do not take into account the level of detail regarding current or future events, and therefore fail to adequately address user's needs.

BRIEF SUMMARY

In one aspect, a computer-implemented method for determining a customized loan for a user may be provided. The method may include (1) receiving a request for a loan amount for a loan term; (2) calculating a statistical risk of default on the requested loan; (3) calculating a loan rate for the requested loan amount and loan term based upon the statistical risk of default; (4) receiving user input data including (i) demographic data for the user, and (ii) user behavioral data associated with the user; (5) calculating an adjusted statistical risk of default on the requested loan based upon the user input data; (6) adjusting the calculated loan rate for the requested loan to an adjusted loan rate based upon the adjusted statistical risk of default; and/or (7) presenting the adjusted loan rate to the user. The method may include additional, less, or alternate components, including those discussed elsewhere herein.
In yet another aspect, a computer system for predicting when a user requires a loan and presenting the user with a loan offer may be provided. The system may include (1) a client device (or mobile device) associated with a user, which may be configured to transmit a request for a loan amount for a loan term via wireless communication or data transmission over one or more radio links or wireless communication channels; and (2) one or more back-end components configured to (a) receive the request for the loan amount for the loan term; (b) calculate a statistical risk of default on the requested loan; (c) calculate a loan rate for the requested loan amount and loan term based upon the statistical risk of default; (d) receive user input data including (i) demographic data for the user, and (ii) user behavioral data; (e) calculate an adjusted statistical risk of default on the requested loan based upon the user input data; (f) adjust the calculated loan rate for the requested loan to an adjusted loan rate based upon the adjusted statistical risk of default; and/or (g) transmit a notification including the adjusted loan rate to the client device via wireless communication or data transmission over one or more radio links or wireless communication channels. The system may include additional, less, or alternate components, including those discussed elsewhere herein.
In still another aspect, a computer-implemented method for determining a customized loan for a user may be provided. The method may include (1) receiving a requested loan amount for a loan term, such as from user mobile or computing device via wireless communication or data transmission over one or more radio links or wireless communication channels; (2) calculating a statistical risk of default on the requested loan; (3) calculating a loan rate for the requested loan amount and loan term based upon the statistical risk of default; (4) receiving user input data including (i) demographic data for the user, and (ii) user behavioral data; (5) calculating an adjusted statistical risk of default on the requested loan based upon the user input data; (6) adjusting the calculated loan rate for the requested loan to an adjusted loan rate based upon the adjusted statistical risk of default; and/or (7) presenting the adjusted loan rate to the user, such as by transmitting the adjusted loan rate in an electronic message to the user's mobile or computing device via wireless communication or data transmission over one or more radio links or wireless communication channels. The method may include additional, less, or alternate components, including those discussed elsewhere herein.
Advantages will become more apparent to those of ordinary skill in the art from the following description of the preferred aspects which have been shown and described by way of illustration. As will be realized, the present aspects may be capable of other and different aspects, and their details are capable of modification in various respects. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a block diagram of an exemplary loan customization system 100 in accordance with one aspect of the present disclosure;

FIG. 2 illustrates exemplary user profiles 200 in accordance with one aspect of the present disclosure;

FIG. 3 illustrates exemplary logic diagrams 300 indicating several different example scenarios that may impact the initial calculation of loan specifics for a certain type of loan requested by a user in accordance with one aspect of the present disclosure; and

FIG. 4 illustrates an exemplary computer-implemented method flow 400 in accordance with one aspect of the present disclosure.

The Figures depict aspects of the present invention for purposes of illustration only. One skilled in the art will readily recognize from the following discussion that alternate aspects of the structures and methods illustrated herein may be employed without departing from the principles of the invention described herein.

DETAILED DESCRIPTION

The present aspects discussed herein may further relate to, inter alia, using cognitive computing and/or predictive modeling (and/or machine learning techniques) to provide a customized loan on a per-user basis. This customized loan may be applicable to the various loan offers upon the determination that a user is actively seeking a loan, as discussed immediately above. Additionally or alternatively, this customized loan may apply to separate loans solicited by the user in various ways (e.g., online, in person, over the phone, etc.). To facilitate the calculation of customized loan specifics, one or more back-end components may collect user input data, which may include the aforementioned information and additionally or alternatively include data such as credit score data, credit report data, and income, from which a statistical risk of default on a particular loan may be calculated.
In the context of determining statistical risk of default, collected behavioral data may include data that is leveraged to predict changes in the user's statistical risk of default over the life of a proposed loan term. For example, the behavioral data may indicate that the user is currently attending classes for a particular field of study. This data may then be used to adjust the statistical risk of default and the calculated loan rate accordingly to present the user with a tailored loan rate or other loan specifics based upon that user's profile.

System Overview

FIG. 1 is a block diagram of an exemplary loan customization system 100 in accordance with one aspect of the present disclosure. In the present aspect, loan customization system 100 may include one or more client devices 102, a personalized loan engine 120, one or more financial institutions 150, and/or a communication network 116. Loan customization system 100 may include additional, less, or alternate components, including those discussed elsewhere herein.
For the sake of brevity, loan customization system 100 is illustrated as including a single client device 102, a single personalized loan engine 120, two financial institutions 150, and a single communication network 116. However, the aspects described herein may include any suitable number of such components. For example, personalized loan engine 120 may communicate with several client devices 102, each of which may be operated by a separate user, to receive data from each separate client device 102 and/or to transmit notifications to each separate client device 102, as further discussed herein. To provide another example, personalized loan engine 120 may receive data from one or more client devices 102 such that a user profile for each user may include data received from each user's respective client device. To provide yet another example, client device 102 may represent one client device from several different client devices for the same user or for different users. For example, client device 102 may represent a user's smartphone as well as a user's desktop computer, each of which may collect and transmit data to one or more financial institutions 150 and/or personalized loan engine 120, as further discussed below.
Communication network 116 may be configured to facilitate communications between one or more client devices 102, one or more financial institutions 150, and/or personalized loan engine 120 using any suitable number of wired and/or wireless links, such as links 117.1-117.3, for example. For example, communication network 116 may include any suitable number of nodes, radio frequency links, wireless or digital communication channels, additional wired and/or wireless networks that may facilitate one or more landline connections, internet service provider (ISP) backbone connections, satellite links, public switched telephone network (PSTN), etc.
To facilitate communications between the various components of loan customization system 100, the present aspects include communication network 116 being implemented, for example, as a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), or any suitable combination of local and/or external network connections. To provide further examples, communications network 116 may include wired telephone and cable hardware, satellite, cellular phone communication networks, base stations, macrocells, femtocells, etc. In the present aspects, communication network 116 may provide one or more client devices 102 with connectivity to network services, such as Internet services, for example, and/or support application programming interface (API) calls between one or more client devices 102, one or more financial institutions 150, and/or personalized loan engine 120.
Client device 102 may be configured to communicate using any suitable number and/or type of communication protocols, such as Wi-Fi, cellular, BLUETOOTH, NFC, RFID, etc. For example, client device 102 may be configured to communicate with communication network 116 using a cellular communication protocol to send data to and/or receive data from one or more financial institutions 150 and/or personalized loan engine 120 via communication network 116 using one or more of radio links or radio frequency links 117.1-117.3 or wireless communication channels.
In various aspects, client device 102 may be implemented as any suitable communication device. For example, client device 102 may be implemented as a user equipment (UE) and/or client device, such as a smartphone, for example. To provide additional examples, client device 102 may be implemented as a personal digital assistant (PDA), a desktop computer, a tablet computer, a laptop computer, a wearable electronic device, etc.
As further discussed below, data collected and/or transmitted by client device 102 to one or more financial institutions 150 and/or personalized loan engine 120 may include, for example, any suitable or relevant information used by personalized loan engine 120 to track a location of client device 102 and/or to track other types of information about users associated with client device 102 to anticipate whether the user associated with client device 102 may require a loan. Additionally or alternatively, data collected and/or transmitted by client device 102 to one or more financial institutions 150 and/or personalized loan engine 120 may include data used by personalized loan engine 120 to preapprove a user for a loan and/or to calculate an adjusted loan rate or other loan specifics for a particular user.
For example, with user permission, such as user opt-in to a rewards or other program that provides financial benefits or cost savings to the user, client device 102 may collect, monitor, store, and/or transmit demographic information, data indicative of the user's behavior such as spending habits, where the user has shopped physically and/or online, data indicative of certain life events, financial information such as account balances of one or more users associated with client device 102, online web browsing history, life event data, etc. This data is discussed in more detail below with reference to FIG. 2.
Furthermore, data received by client device 102 from personalized loan engine 120 may include any suitable information used to notify the user of relevant loan offers, the result of a loan preapproval, whether the user qualifies for a particular loan product and, if so, the particular loan products the user may qualify for. Additionally or alternatively, data received by client device 102 from personalized loan engine 120 may facilitate providing user notifications regarding specific loan terms and/or loan rates, whether a particular loan includes personalized rates or other terms, and the specific details about how the customized loan details were adjusted for that user.
For example, if it is determined by personalized loan engine 120 that a user associated with client device 102 is likely to require a loan, then client device 102 may display a notification transmitted via personalized loan engine 120 regarding loan offers of a specific loan type and amount before the user has obtained the loan. To provide another example, if a user applies for a loan via client device 102 or in another manner, personalized loan engine 120 may calculate user-specific loan specifics that take into account changes in the user's statistical risk during the loan term, which is further discussed below. Assuming the loan is approved, personalized loan engine 120 may adjust the initial loan specifics based upon this additional information and transmit this information to client device 102, which may in turn display a suitable notification including the details of the personalized loan specifics.

Detailed Operation of Loan Customization System

In the present aspects, client device 102 may include one or more processors 104, a communication unit 106, a user interface 108, a display 110, a location acquisition unit 112, and a memory unit 114.
Communication unit 106 may be configured to facilitate data communications between client device 102 and one or more of communication network 116, one or more financial institutions 150, and/or personalized loan engine 120 in accordance with any suitable number and/or type of communication protocols. In the present aspects, communication unit 106 may be configured to facilitate data communications based upon the particular component and/or network with which client device 102 is communicating.
Such communications may facilitate the transmission of collected data from client device 102 that is utilized by personalized loan engine 120 to provide loan customization and/or to predict when a user associated with client device 102 is likely in the market for a new loan, as further discussed herein. In the present aspects, communication unit 106 may be implemented with any suitable combination of hardware and/or software to facilitate this functionality. For example, communication unit 106 may be implemented with any suitable number of wired and/or wireless transceivers, network interfaces, physical layers (PHY), ports, antennas, etc.
User interface 108 may be configured to facilitate user interaction with client device 102. For example, user interface 108 may include a user-input device such as an interactive portion of display 110 (e.g., a “soft” keyboard displayed on display 110), an external hardware keyboard configured to communicate with client device 102 via a wired or a wireless connection (e.g., a BLUETOOTH keyboard), an external mouse, or any other suitable user-input device.
Display 110 may be implemented as any suitable type of display that may facilitate user interaction, such as a capacitive touch screen display, a resistive touch screen display, etc. In various aspects, display 110 may be configured to work in conjunction with user-interface 108 and/or one or more processors 104 to detect user inputs upon a user selecting a displayed interactive icon or other graphic, to identify user selections of objects displayed via display 110, to display notifications regarding available loan offers, preapproval results, the terms of a particular customized loan, which may be received, for example, via personalized loan engine 120, etc.
Location acquisition unit 112 may be implemented as any suitable device configured to generate location data indicative of a current geographic location of client device 102. In one aspect, location acquisition unit 102 may be implemented as a satellite navigation receiver that works with a global navigation satellite system (GNSS) such as the global positioning system (GPS) primarily used in the United States, the GLONASS system primarily used in the Soviet Union, the BeiDou system primarily used in China, and/or the Galileo system primarily used in Europe.
Location acquisition unit 112 and/or one or more processors 104 may be configured to receive navigational signals from one or more satellites and to calculate a geographic location of client device 102 using these signals. Location acquisition unit 112 may include one or more processors, controllers, or other computing devices and memory to calculate the geographic location of client device 102 without one or more processors 104. Alternatively, location acquisition unit 112 may utilize components of one or more processors 104. Thus, one or more processors 104 and location acquisition unit 112 may be combined or be separate or otherwise discrete elements.
One or more processors 104 may be implemented as any suitable type and/or number of processors, such as a host processor for the relevant device in which client device 102 is implemented, for example. One or more processors 104 may be configured to communicate with one or more of communication unit 106, user interface 108, display 110, location acquisition unit 112, and/or memory unit 114 to send data to and/or to receive data from one or more of these components.
For example, one or more processors 104 may be configured to communicate with memory unit 114 to store data to and/or to read data from memory unit 114. In accordance with various embodiments, memory unit 114 may be a computer-readable non-transitory storage device, and may include any combination of volatile (e.g., a random access memory (RAM)), or a non-volatile memory (e.g., battery-backed RAM, FLASH, etc.). In the present aspects, memory unit 114 may be configured to store instructions executable by one or more processors 104. These instructions may include machine readable instructions that, when executed by one or more processors 104, cause one or more processors 104 to perform various acts.
In the present aspects, personalized loan application 115 is a portion of memory unit 114 configured to store instructions, that when executed by one or more processors 104, cause one or more processors 104 to perform various acts in accordance with applicable aspects as described herein. For example, instructions stored in personalized loan application 115 may facilitate one or more processors 104 performing functions such as periodically reporting and/or transmitting the location of client device 102 as part of a running background process (or causing location acquisition unit 112 to do so), collecting various types of data, sending various types of data to one or more financial institutions 150 and/or personalized loan engine 120, receiving data and/or notifications from one or more financial institutions 150 and/or personalized banking engine 120, displaying notifications and/or other information using data received via one or more financial institutions 150 and/or personalized banking engine 120, etc.
In some aspects, personalized loan application 115 may reside in memory unit 114 as a default application bundle that may be included as part of the operating system (OS) utilized by client device 102. But in other aspects, personalized loan application 115 may be installed on client device 102 as one or more downloads, such as an executable package installation file downloaded from a suitable application source via a connection to the Internet or other suitable device, network, external memory storage device, etc.
For example, personalized loan application 115 may be stored in any suitable portions of memory unit 114 upon installation of a package file downloaded in such a manner. Examples of package download files may include downloads via the iTunes store, the Google Play Store, the Windows Phone Store, a package installation file downloaded from another computing device, etc. Once downloaded, personalized loan application 115 may be installed on client device 102 as part of an installation package such that, upon installation of personalized loan application 115, memory unit 114 may store executable instructions such that, when executed by one or more processors 104, cause client device 102 to implement the various functions of the aspects as described herein.
The user may initially create a user profile upon first launching personalized loan application 115, through a registration process via a website, over the phone, etc. This user profile may include, for example, the customer's demographic information or any suitable information that may be useful in facilitating various portions of the aspects as described herein. For example, upon installing and launching personalized loan application 115 on client device 102, a user may be prompted to enter login information and/or complete an initial registration process to create a user profile with a lender or other relevant party associated with or otherwise affiliated with personalized loan engine 120.
Additionally or alternatively, personalized loan application 115 may periodically request data directly from the user as opposed to collecting data in a passive manner. For example, personalized loan application 115 may request certain types of information from the user as part of one or more surveys and/or questionnaires. This information may be requested, for example, upon an initial registration and/or at any suitable time after the initial registration. In this way, a user may be asked for certain types of information that may be difficult to obtain through third party data providers and/or public records. For example, a user may be asked about the ages of his family members, his current income level (or where his earnings fit within a range of earnings), his hobbies and interests, whether he is planning to attend a university or other training program in the near future and, if so, in what field of study, whether he intends to make a large purchase in the near future, etc.
An initial registration process may additionally or alternatively include, for example, obtaining the user's consent to track her location or to otherwise collect and utilize other types of data to provide the various loan customization and/or prediction services as discussed herein. For example, a user may opt in to allow personalized loan engine 120 to track and/or receive the user's financial account data including financial transactions, spending history, credit card balances, web browsing history, and/or account balances associated with the user's financial accounts at one or more financial institutions 150, to opt in to a system whereby the user's spending habits are tracked and/or spending data at various retailers is collected, to provide authorization (e.g., online login credentials) to access the user's accounts held at one or more financial institutions 150, etc. In return, the user may be presented with various offers or customized financial products or loan offers, or other products discussed herein.
In the present aspects, personalized loan application 115 may provide different levels of functionality based upon options selected by a user and/or different implementations of personalized loan application 115. For example, in some aspects, personalized loan application 115 may facilitate client device 102 working in conjunction with one or more financial institutions 150 and/or personalized banking engine 120 to facilitate receiving notifications for offers regarding loans anticipated by personalized loan engine 120. To provide another example, other aspects include personalized loan application 115 facilitating client device 102 working in conjunction with one or more financial institutions 150 and/or personalized banking engine 120 to collect and transmit data to one or more financial institutions and/or personalized loan engine 120. This may, for example, facilitate the customization of loan specifics for loans anticipated by personalized loan engine 120 and/or for a separate loan that may be, for example, be solicited by the user via client device 102 or in another suitable manner.
One or more financial institutions 150 may include any suitable number and/or type of financial institutions that hold and/or are associated with various financial accounts. For example, one or more financial institutions 150 may include banks, creditors, lenders, and/or brokers. One or more users (e.g., a user associated with client device 102) may hold one or more accounts with one or more financial institutions 150 such as checking accounts, savings accounts, credit accounts, lines of credit, loan accounts, charge accounts, money market accounts, brokerage accounts, etc. These accounts may be held at a single institution or spread out across several different financial institutions.
In one aspect, financial accounts held at one or more financial institutions 150 may be accessible via a secure connection to communication network 116, for example, by client device 102 and/or personalized loan engine 120. For example, one or more financial institutions 150 may provide online services that allow a user to access her accounts using client device 102 and/or another suitable computing device. Upon receipt of a valid authenticated request for financial data, one or more financial institutions 150 may transmit financial data to client device 102 and/or personalized loan engine 120. Examples of the financial data transmitted by one or more financial institutions 150 may include financial transaction data indicating previous credits and debits to a user's accounts, a current account balance, loan payoff balances, credit report history, credit scores, credit card utilization, derogatory credit marks, spending data such as the time, amount, and specific merchant for which previous account debits and/or charges were made, whether the user has previously defaulted on a particular loan, etc.
Personalized loan engine 120 may be affiliated or otherwise associated with one or more parties, which may be the same party or a different party than those affiliated with one or more financial institutions 150. Personalized loan engine 120 may include any suitable number of components configured to receive data from and/or send data to one or more of client devices 102 and/or one or more financial institutions 150 via communication network 116 using any suitable number of wired and/or wireless links. In various embodiments, personalized loan engine 120 may constitute a portion of (or the entirety of) one or more back-end components, and may be configured (alone or in conjunction with other back-end components) to execute one or more applications to perform one or more functions associated with the various aspects as discussed herein.
For example, as shown in FIG. 1, personalized loan engine 120 may communicate with one or more external computing devices such as servers, databases, database servers, web servers, etc. The present aspects include personalized loan engine 120 working in conjunction with any suitable number and/or type of back-end components to facilitate the appropriate functions of the aspects as described herein.
In the present aspects, personalized loan engine 120 may be implemented, for example, as any suitable number and/or type of servers configured to access data from one or more additional data sources and/or store data to one or more storage devices. For example, as shown in FIG. 1, additional data sources 170 may represent data that is made accessible by personalized loan engine 120 from any number N of data sources, such as third-party data providers or other data sources in addition to and/or including one or more financial institutions 150. This additional data may be utilized by personalized loan engine 120 to build a personalized user profile for each user, which may be stored in one or more suitable storage units (e.g., storage unit 180) and include other types of data, as further discussed below. Personalized loan engine 120 may access each user's profile as part of the execution of one or more cognitive computing and/or predictive modeling algorithms to calculate a statistical likelihood that a user will default on a particular loan and/or to determine a specific loan type and accompanying loan specifics for a particular user based upon that user's profile.
To provide some illustrative examples, additional data sources 170 may include, again with user permission, data mined from social media and/or user's web browsing habits (e.g., search terms and websites), data provided by various retailers, demographic data associated with other users in a particular region and/or associated with a particular retailer, income levels of various users, where different users commonly shop, assets associated with various users such as the cars each user owns, mortgage loan information associated with various users, how much users typically spend at various retailers, etc. To provide further examples, additional data sources 170 may include specific demographic information for a particular user and his family members, such as the age and gender of each child, for example.
To provide even more examples, additional data sources 170 may include data indicative of various user life events such as getting married, a child about to attend (or currently attending) college, paying off a previous loan, receiving a settlement or inheritance, etc. Still further, additional data sources 170 may include user data such as spending habits or detailed information such as a college or university the user may be attending, courses being taken by the user, a college major or other focused field of study, etc. Personalized loan engine 120 may utilize any portion of such data (as well as data from other data sources) as input to one or more cognitive computing and/or predictive modeling algorithms to perform the predictions and statistical calculations as discussed herein, some examples of which are further discussed below.
Personalized loan engine 120 may be implemented as any suitable number of servers that are configured to generate and/or store various user profiles in storage unit 180. Each user's profile may include, for example, an aggregation of aforementioned data and/or any other suitable data that may be utilized to calculate a statistical risk of a user defaulting on a loan and/or data which may be used to calculate loan specifics based upon a user's profile.
For example, a user's profile may include demographic data, data submitted by a user as part of an initial registration process, data submitted by a user in response to solicited surveys, spending data, financial data (e.g., credit card utilization, credit card balances, bank account balances, etc.), credit report data, credit scores, family or individual income, etc. Each user's profile may be stored in storage unit 180 in any suitable manner such that personalized banking engine 120 may access each user's profile and correlate each user to his or her profile.
To provide an illustrative example, storage unit 180 may include a number of user profiles organized in accordance with suitable type of information to uniquely identify each particular user so that each user may later be matched to her profile stored in storage unit 180. For example, each user's profile may be identified by a username that is used by one or more users in accordance with personalized loan application 115, a first and last name of each user, etc. These user profiles are discussed in further detail below with reference to FIG. 2.
In the present aspects, personalized loan engine 120 may include one or more processors 122, a communication unit 124, and a memory unit 126. One or more processors 122, communication unit 124, and memory unit 126 may perform substantially similar functions as one or more processors 104, communication unit 106, and memory unit 114, respectively, of client device 102. Therefore, only differences between these components will be further discussed herein.
Of course, differences between components of personalized loan engine 120 and client device 102 may be owed to differences in device implementation rather than the functions performed by each individual component. For example, if personalized loan engine 120 is implemented as a server whereas client device 102 is implemented as a personal computing device, one or more processors 122 may have more processing power (e.g., a faster processor, more cores, etc.) than one or more processors 104, although one or more processors 122 may perform similar functions as one or more processors 104 (e.g., executing instructions stored in memory to perform various acts, processing data, etc.).
In the present aspects, personalized loan engine 120 may be configured to send or otherwise transmit various types of notifications and/or inquiries to client device 102. These notifications and/or inquiries may be sent, for example, from personalized loan engine 120 via communication unit 124, and may include any suitable type of data transmissions. For example, personalized loan engine 120 may transmit appropriate notifications and/or inquiries via emails, text messages, push notifications, etc., to client device 102
Again, in various aspects, personalized loan engine 120 may acquire data from various sources to facilitate the various aspects described herein. Some of these sources may include data from secure connections or may otherwise require secured or authorized access. For example, one or more financial institutions 150 may provide access to one or more user's financial account data via any suitable authentication techniques, such as via a secure connection, password authentication, public and/or private key exchanges, biometric identification, etc.
Therefore, in the present aspects, personalized loan engine 120 may, when appropriate, implement any suitable techniques to obtain information in a legal and technically feasible manner. For example, as discussed above, a user may setup an account and/or profile with a third party associated with personalized loan engine 120. As discussed above, the user may then opt in to data collection via the various data sources that are to be collected via personalized loan engine 120. The user may additionally or alternatively provide authentication information for each account and/or data source for which data is to be accessed, collected, tracked, monitored, etc., such that personalized loan engine 120 may obtain any suitable type of data to carry out the aspects described herein.

Cognitive Computing

Cognitive computing may refer to systems that learn at scale, reason with purpose, and/or interact with humans naturally. Rather than being explicitly programmed, cognitive computing systems may learn and reason from their interactions and from their experiences with their environment. As opposed to traditional computing systems, which are deterministic, cognitive computing systems are probabilistic. In other words, cognitive computing systems generate not just answers to numerical problems, but hypotheses, reasoned arguments, and recommendations about more complex and meaningful bodies of data. Cognitive computing systems may advantageously interpret and utilize data that is typically referred to as being unstructured in nature. This allows such systems to keep pace with the volume, complexity, and unpredictability of information and systems in the modern world. To do so, cognitive computing systems attempt to augment the reasoning and thought processes of the human brain.
Therefore, in various aspects, personalized loan engine 120 may be implemented as a computing device (or a constituent part of one or more computing devices) that is/are configured to process data in accordance with one or more cognitive computing techniques. For example, personalized loan engine 120 may be implemented as one or more nets or nodes of an artificial neural network system and/or other suitable system that models and/or mimics the reasoning and processing of the human brain. Thus, cognitive computing and predictive modeling application 127 may include one or more machine learning algorithms, code, logic, and/or instructions to facilitate the behavior, functionality, and/or processing of a cognitive computing system.
In the present aspects, cognitive computing and predictive modeling application 127 may include any suitable combination of functions as discussed herein. For example, as shown in FIG. 1, cognitive computing and predictive modeling application 127 may include a data aggregation module 129 and a personalized loan module 133. These modules are for illustrative purposes and represent examples of some of the functionality that may be performed by personalized loan engine 120 in accordance with a cognitive computing-based system. However, aspects include cognitive computing and predictive modeling application 127 including additional, less, or alternate actions, including those discussed elsewhere herein. Furthermore, aspects include cognitive computing and predictive modeling application 127 implementing traditional, non-cognitive computing processes.
In one aspect, data aggregation module 129 may be a portion of memory unit 126 configured to store instructions, that when executed by one or more processors 122, cause one or more processors 122 to perform various acts in accordance with applicable embodiments as described herein. In the present aspects, instructions stored in data aggregation module 129 may facilitate one or more processors 122 performing functions such as mining data for any suitable number of users. For example, instructions stored in data aggregation module 129 may facilitate personalized loan engine 120 providing the requested authorization to one or more financial institutions 150 and/or additional data sources 170 as needed and to receive data from one or more of these sources.
In the present aspects, instructions stored in data aggregation module 129 may facilitate personalized loan engine 120 aggregating and organizing received data into one or more user profiles, which may then be stored in storage unit 180. Again, these user profiles may be associated with each user and organized such that data contained as part of each user's profile may be associated with each user. For example, each user's username and/or other suitable identifying information may be stored as part of a user profile that includes an aggregation of all data received for that particular user from one or more data sources, as previously discussed.
In the present aspect, personalized loan module 133 is a portion of memory unit 126 configured to store instructions, that when executed by one or more processors 122, causes one or more processors 122 to perform various acts in accordance with applicable aspects as described herein. In the present aspects, instructions stored in personalized loan module 133 may facilitate one or more processors 122 performing functions such as adjusting a loan rate, adjusting the length of a loan, and/or adjusting other specifics associated with a particular loan based upon various sources of data. For example, personalized loan module 133 may facilitate the calculation of a customized loan terms based upon an analysis of data that may determine not only a present statistical risk of a user defaulting on a loan, but an adjusted statistical risk that compensates for changes in user information or user behavior over the term of the loan.
That is, a user may request a loan in person or, for example, submitting a request from a suitable computing device (e.g., client device 102). This request may include, for example, a loan amount, a type of loan, a loan term, etc. In such a case, an initial statistical risk of default on the requested loan may be calculated in accordance with any suitable techniques, which may include, for example, any suitable number and/or type of user information contributing to the calculation of an initial statistical risk of default. This may include, for example, the utilization of information such as such as credit history, credit scores, previous loan defaults, family income, debt-to-income ratio, etc. In an aspect, this initial statistical risk of default may be calculated partially or entirely using traditional techniques of risk analysis. In other aspects, this initial statistical risk of default may utilize additional sources of data from the user's profile that would not ordinarily be used in traditional risk analyses, which is further refined with even more user input data, as discussed below.
In any event, continuing this example, once an initial statistical risk of default is calculated, a loan rate (i.e., interest rate) may be calculated that provides users with higher risks of default higher loan rates, and more favorable interest loan rates for lower risk clients. In the present aspects, personalized loan module 133 may facilitate the adjustment of the initial statistical risk of default and, in turn, an adjustment of the initially calculated loan rate.
To provide an additional illustrative example, a user may request a 30-year mortgage loan for $200,000. The user may have a relatively low risk of default based upon his credit history, but may have a relatively high debt-to-income ratio. Continuing this example, personalized loan engine 120 may determine from an analysis of demographic data and/or behavioral data (e.g., received via client device 102 and/or data sources 170) that the client is currently attending a university majoring in electrical engineering. Using this information, the client's potential future earnings may be calculated such that, in the next 4 years, the client's income will increase and his debt-to-income ratio will likely decrease. Because this change will happen early within the 30-year life of the loan, this potential future income may more accurately reflect the future risk of the client defaulting on the requested mortgage loan. Thus, in the present aspects, the initially calculated statistical risk of default and initially calculated loan rate may be adjusted to consider both the client's current earnings as well as the client's potential future earnings during the life of the loan. In this way, the present aspects allow for a customized loan to be presented to the client with a loan rate lower than what would otherwise be available.
In the present aspects, the user's potential future earnings may be calculated in any suitable manner to accurately determine this information. For example, personalized loan engine 120 may identify other users with a demographic profile that matches the user's demographic profile. This may include, for example matching key components of user demographic profiles that are particularly well correlated to earnings (e.g., region and age). From these matched other users, the present aspects include further identifying users who have completed the same training programs or college courses that the user is currently participating—in this case electrical engineering. Personalized loan engine 120 may then average the salaries associated with each of these other users and project (e.g., using inflation prediction models) what this average salary will be at a point in the future within the term of the loan (e.g., four years from when the client is applying). The client's potential future earnings may then be set as the calculated projected average salary of the other users in four years.
Although the aforementioned examples focus on the adjustment of an initially calculated loan rate, the present aspects encompass the use of predicted potential earnings or any other future factors that impact risk to adjust any suitable loan specifics to provide a customizes loan on a per-user basis. For example, the loan term may be adjusted if this results in a higher payment but a lower overall cost for the user when the result of the adjustment yields an acceptable level of adjusted statistical risk. To provide another example, the type of loan may be adjusted to consider future earnings. In the case of a mortgage loan, this could include, for example, adjusting a mortgage loan from an initial 30 year loan to a 5/1 arm when the user's potential future earnings indicate that after 5 years the statistical risk of default based upon any future rate changes is acceptable.
Again, the adjusted loan rate or other loan specifics may be based upon predicted or future risk in addition to current statistical risk. Of course, the loan interest rate may be adjusted up or down based upon the adjusted statistical risk of default when this additional information is considered, and the adjusted loan interest rates may be calculated in any suitable manner. For example, when a user's potential future earnings are greater than the user's current earnings in excess of a threshold amount, the calculated loan rate may be adjusted to a lower loan rate, and vice-versa. Continuing this example, the calculated loan rate may be adjusted to a lower loan rate by an amount that is proportional to an amount in which the user's potential future earnings exceed the user's current earnings. That is, for every ten percent the user's future earnings exceed the user's current earnings, the interest rate may be adjusted another tenth of a point downward. In this way, the present aspects calculate statistical risks based upon both present and future information to provide customized loan terms for each user's specific present and future choices, behaviors, and demographics.
Once the loan specifics are calculated, the present aspects include personalized loan engine 120 transmitting one or more notifications to a suitable computing device associated with the user (e.g., client device 102) via communication unit 124, network 116, and links 117.1 and 117.3, for example. The computing device may then display these notifications and/or receive user input from the user in response to receiving these notifications. Furthermore, aspects include the user's computing device displaying the adjusted loan rate and/or other loan specifics that have been calculated by personalized loan engine 120 on a display associated with the computing device (e.g., display 110) in response to the user responding to the notification. For example, client device may transmit data indicating that the user would like to view loan specifics that have been calculated for a specific loan. Upon receiving this data, personalized loan engine 120 may transmit the loan specifics to client device 102, where they are then displayed. In this way, a user may request a specific type of loan and then view the loan specifics associated with that loan using the same computing device.

Exemplary Calculations for Predicting Loan Needs Using Cognitive Computing

FIG. 2 illustrates exemplary user profiles 200 in accordance with one aspect of the present disclosure. As shown in FIG. 2, user profiles 200 are an example of the various types of data that may be stored in any suitable number of storage units or databases (e.g., storage unit 180, as shown in FIG. 1). User profiles 200 may be generated, organized, modified, and/or accessed via a personalized loan engine, such as personalized loan engine 120, for example, as shown in FIG. 1.
As shown in FIG. 2, user profiles 200 may include a number of different types of collected data associated with a number of individual users A-D. FIG. 2 illustrates four exemplary user profiles and five different types of data associated with each user profile. However, the present aspects include user profiles 200 including any suitable number of user profiles for any suitable number of different users, which may be associated with any suitable number and/or type of data.
User profiles 200 illustrate a number of different types of data associated with each user. The different types of data aggregated as part of each user's financial profile may be collectively referred to as “user input data,” although the user input data may include information about the user as well as other people, such as the user's family members. Again, this user input data may be used in accordance with the present aspects to perform calculations and predictions regarding whether a particular user will soon require a loan and/or the calculation of specific customized loan specifics.
As shown in FIG. 2, each user's financial profile contains different types of information, and portions of each respective type of information may represent user inputs for cognitive computing and predictive modeling application 127. Inputs (a1-a6) may represent each user's demographic data, such as the user's individual age and/or the ages of each member of the user's family (e.g., their birthdates), gender, marital status, household size, whether the user owns a home or rents, the user's ethnicity, the ethnicity of members of the user's family, etc.
Inputs (b1-b3) may represent different types of financial data such as, for example, the user's current individual and/or family annual income, the user's potential individual and/or family annual income (which may be calculated from other portions of the user's financial profile, as previously discussed), the user's credit information such as credit card utilization, debt-to-income ratios, credit scores, credit reports, etc.
Inputs (c1-c3) may represent various types of behavioral data that indicate the user's previous and likely future choices, behaviors, and/or actions. For example, behavioral data may represent online data, such as online search terms, the content of email such as various identified key words and their frequency of use, various websites visited by the user, etc.
To provide an illustrative example, behavioral data may indicate that the user recently researched different makes and models of cars by visiting several car review websites, by entering online search terms for specific types of vehicles, and/or by visiting automotive manufacturer websites. To provide additional illustrative examples, the behavioral data may indicate that a user has visited different lending websites seeking specific types of loans. To provide yet another illustrative example, the behavioral data may indicate that the user has visited several real-estate based websites looking to purchase a home and the range of home values searched. To provide even further illustrative examples, behavioral data may indicate that a user has completed an online application for a university, has researched options for student loans, and/or has visited certain college or university websites.
Inputs (d1-d6) may represent various types of life event data that may indicate if and when the user is ready to acquire a new loan. The life event data may include different types of data from public records, data submitted and/or collected from the user, and/or data collected via other third party sources. As shown in FIG. 2, the life event data may include information such as birth records (e.g., birth certificates recently recorded), marriage certificates, an indication that a child has started college, a determination of whether a child is approaching legal driving age, whether the user has recently purchased a new home, an indication that a user's current home is a particular age that may require repairs (and thus the user may be potentially interested in a home equity loan), etc.
Inputs (e1-e2) may represent different types of location data that may indicate the user's current geographic location and/or a history of the user's previous locations. In the present aspects, this location data may be analyzed and/or referenced to other locations known to be associated with a user actively seeking or otherwise likely to require a monetary loan. To provide an illustrative example, the location data may indicate that the user recently visited physical geographic locations associated with car dealerships, college campuses, physical banks or lender locations, high-end retailers, etc.
Again, in the present aspects, personalized loan engine 120 may actively collect, aggregate, and/or monitor data representing user inputs for each user's profile to calculate adjusted loan terms, as discussed herein. To perform these tasks, the present aspects include personalized loan engine 120 analyzing the user input data in accordance with any suitable predictive modeling and/or cognitive computing techniques.
For example, assume that it is identified that user A is attending college majoring in electrical engineering, and will graduate in 1 year. Further assume that users B-D have a demographic profile matching that of user A. For example, this may be determined when users B-D have similar demographic data such that a threshold number of inputs a1-a6 match or match within a certain threshold, those of user A. In such a scenario, the present aspects include personalized loan engine 120 calculating an initial loan rate, term, and amount for user A to purchase a new home using user A's current income level (i.e., input b1). Personalized loan engine 120 may then average the corresponding inputs b1 from each of users B-C and use this averaged income value as user A's potential future earnings, which may compensate for inflation or market factors in the next year. Personalized loan engine 120 may then adjust the loan term, amount, rate, etc., based upon both user A's current income (b1) and user A's potential future income (b2).
In the present aspects, this adjustment may be made in accordance with any suitable manner. For example, personalized loan engine 120 may calculate a confidence level of this prediction by using a statistical probability or likelihood of the calculated potential earning being correct. In such a case, more weight may be given (e.g., the rate may be reduced by a larger amount) for calculated potential earnings associated with a higher likelihood of being correct, while less confident calculations may result in less adjustments to the initial loan term calculations.

Exemplary Logic Diagram for Different Scenarios Impacting the Adjustment of Loan Specifics

FIG. 3 illustrates exemplary logic diagrams 300 indicating several different example scenarios that may impact the initial calculation of loan specifics for a certain type of loan requested by a user in accordance with one aspect of the present disclosure. As shown in FIG. 3, each of examples 1-3 is based upon different portions of user input data stored as part of user profiles associated with three different users A-C. The user profiles shown in FIG. 3 may be implementations of user profiles 200, for example, as shown in FIG. 2.
As discussed above, aspects include portions of a user's profile being analyzed to determine future events and/or a statistical likelihood that a user will default on a particular loan based upon changes in the user's behavior, as indicated by the user's input data, over the life of the loan. For example, as shown in Example 1 of FIG. 3, at the time a user A has requested a mortgage loan, user A's profile may reveal that user A was recently admitted to a particular state university and that user A will major in electrical engineering. Using this information, personalized loan engine 120 may determine an initial statistical rate of default on the requested mortgage loan using user input data that does not include the information shown in FIG. 3. For example, the initial statistical rate of default on the requested mortgage loan may be based upon user A's credit history, credit rating, and/or income at the time the loan was requested.
Personalized loan engine 120 may then incorporate the user input data shown in FIG. 3 to adjust the calculated statistical rate of default on the mortgage loan and, in turn, adjust the loan specifics accordingly. For example, because electrical engineering degrees have historically provided users with a higher income upon graduation, personalized loan engine 120 may utilize this additional information to determine that in the near future (e.g., about 4-5 years from the request, which is early in the life of most mortgage loans) that user A's income will be greater than it is today. As a result, the statistical risk of default will likewise be lower in 4 years, and the adjusted loan specifics may reflect this decrease in risk.
To provide another example, Example 2 illustrates that at the time a user B has requested a vehicle loan, user B's profile indicates that user B has increased her income over the last 5 years by 7% each year, that user B was married on Aug. 6, 2014, and a new baby was born on May 13, 2015. Again, in this scenario, personalized loan engine 120 may calculate an initial statistical rate of default on the requested automotive loan using user input data that does not include the information shown in FIG. 3. However, personalized loan engine 120 may then use the information shown in FIG. 3 to forecast user B's potential future earnings based upon user B's previous increases in income the last 5 years, which have been consistent and substantial. Furthermore, because personalized loan engine 120 may mimic the logical thought processes of the human brain, other data contained in user B's profile may also be leveraged in conjunction with user B′ forecasted earnings to adjust the statistical rate of default of the loan, and therefore the loan specifics for that loan.
For example, as shown in FIG. 3, user B has been recently married and has a an 18-month old child, both of which strongly correlate to a person's desire to maintain a favorable and responsible credit history in the interest of maintaining financial support for a young family. In other words, not only do does user B's profile indicate a statistical likelihood of her income continuing to increase, but it also indicates behaviors that are relevant to user B's likelihood to not default on a new car loan. Therefore, aspects include personalized loan engine 120 further decreasing the specifics associated with the loan rate (e.g., the interest rates in this example) to better reflect this decrease in default risk for user B.
To provide yet another example, Example 3 illustrates that at the time user C has requested a mortgage loan, that user C's profile indicates that user C has a son who has been recently accepted to a state university and that user C also owns two aging vehicles. Again, in this scenario, personalized loan engine 120 may calculate an initial statistical rate of default on the requested mortgage loan using user input data that does not include the information shown in FIG. 3. However, personalized loan engine 120 may then use data stored in user C's profile to predict that user C will likely need additional loans early in the life of the mortgage loan. Specifically, personalized loan engine 120 may ascertain that user C may soon require a student loan and/or two additional automotive loans due to the age of his current vehicles. In other words, although user C may have a certain debt-to-income ratio at the time the mortgage loan is requested, this metric is likely to increase in the near future. Thus, in contrast to the two previous examples, in this scenario the additional data contained in user C's profile is of a type that is likely to increase the statistical risk of default on the mortgage loan. Therefore, aspects include personalized loan engine 120 increasing the specifics associated with the loan rate (e.g., the interest rates) to better reflect this increase in default risk for user C.
For brevity, the Examples shown in FIG. 3 and previously described illustrate information obtained via one to three user inputs. However, the present aspects include any suitable number of user inputs or other factors being considered as part of the calculation of an initial and/or adjusted rate of default on a particular requested loan, as well as the adjustment to loan specifics based on the adjusted rate of default.

Exemplary Method for Customizing Loan Specifics on a Per-User Basis

FIG. 4 illustrates an exemplary computer-implemented method flow 400 in accordance with an aspect of the present disclosure. In the present aspects, one or more portions of method 400 (or the entire method 400) may be implemented by any suitable device, and one or more portions of method 400 may be performed by more than one suitable device in combination with one another. For example, one or more portions of method 400 may be performed by client device 102 and/or personalized loan engine 120, as shown in FIG. 1. In one embodiment, method 400 may be performed by any suitable combination of one or more processors, instructions, applications, programs, algorithms, routines, etc. For example, method 400 may be performed via or more processors 122 executing instructions stored in cognitive computing and predictive modeling application 127 in conjunction with data collected, received, and/or generated via personalized loan engine 120.
Method 400 may start when one or more processors receive a request for a loan (block 402). The requested loan may have a corresponding loan amount and/or loan term (block 402). The request may be submitted by a user from a client device (e.g., client device 102, as shown in FIG. 1) and, in such a case, may be transmitted or otherwise submitted from the client device to a personalized loan engine (block 402). The requested loan may additionally or alternatively be received in the form of an identified potential loan, such as when a calculated statistical likelihood of a user requiring a loan has been exceeded, as discussed herein (block 402).
Method 400 may include one or more processors calculating a statistical risk of default for the requested loan (block 404). This may include, for example, the statistical risk of default being calculated in accordance with any suitable techniques to adequately assess the risk of the user defaulting on the requested loan (block 404). For example, method 400 may include determining the user's current credit score, debt-to-income ratio, earnings, derogatory marks, payment history, etc. (block 404).
Method 400 may include one or more processors calculating a loan rate for the loan amount (block 406). This may include, for example, calculating a loan rate and/or other specifics of the loan based upon the calculated statistical risk of default (block 404) and/or other factors, as further discussed herein (block 406).
Method 400 may include one or more processors receiving demographic and/or behavioral data as part of user input data (block 408). For example, the user input data may be received from any suitable number and/or type of data sources, such as data collected via client device 102, data collected via one or more financial institutions 150, and/or data collected via one or more additional data sources 170, as shown and discussed with reference to FIG. 1 (block 408). To provide another example, the demographic and/or behavioral data may represent one or more portions of the user input data used to generate a user's profile, as shown and discussed with reference to FIG. 2 (block 408).
Method 400 may include one or more processors calculating an adjusted statistical risk of default on the requested loan based upon the demographic and/or behavioral data (block 410). This may include, for example, adjusting the initially calculated statistical risk of default (block 404) based upon additional factors determined from the demographic and/or behavioral data (block 410). For example, the adjusted statistical risk of default may consider the user's potential future earnings in addition to the user's current earnings, the likelihood that additional loans may be required in the near future, recent additions to the user's family, etc., as discussed herein (block 410).
Method 400 may include one or more processors calculating an adjusted loan rate for the loan amount (block 412). This may include, for example, calculating an adjusted loan rate and other specifics of the loan based upon the adjusted calculated statistical risk of default (block 410) and/or other factors, as further discussed herein (block 412).
Method 400 may include one or more processors presenting the adjusted loan rate and/or other loan specifics (block 412) to the user (block 414). This may include, for example, transmitting a notification to the user's client device (e.g., via text message, email, push notification, etc.), displaying the loan term information locally via a personalized loan engine and/or a computing device in communication with a personalized loan engine, printing out and/or mailing the loan rate and/or other loan specifics, etc. (block 414).

TECHNICAL ADVANTAGES

The aspects described herein may be implemented as part of one or more computer components such as a client device and/or one or more back-end components, such as a personalized loan engine, for example. Furthermore, the aspects described herein may be implemented as part of a computer network architecture and/or a cognitive computing architecture that facilitates communications between various other devices and/or components. Thus, the aspects described herein address and solve issues of a technical nature that are necessarily rooted in computer technology.
For instance, aspects include analyzing various sources of data to calculate the impact of likely future events on a user's statistical rate of default on a loan. In doing so, the aspects overcome issues associated with the inconvenience of manual and/or unnecessary monitoring of user input data by replacing manual procedures with a cognitive-based computing system. Without the improvements suggested herein, additional processing and memory usage would be required to perform such monitoring and/or to calculate the likelihood of various users require a loan.
Furthermore, the embodiments described herein function to personalize loans based upon an analysis of data that facilitates the impact of likely future event. The process improves upon existing technologies by more accurately forecasting such events may concurrently analyzing user input data from a larger number of sources than would be feasible or practical. As a result, the customization of loan specifics (e.g., the loan type, specific loan programs, loan terms, etc.) improves the speed, efficiency, and accuracy in which such calculations could otherwise be performed. Due to these improvements, the aspects address computer-related issues regarding efficiency over the traditional amount of processing power and models used to calculate loan specifics and/or perform data forecasting. Thus, the aspects also improve upon computer technology by requiring less calculations due to the increased efficiency provided, for example, by cognitive computing versus traditional computing techniques. For instance, because cognitive computing may be leveraged to determine whether a user requires a loan, this result reflects an improvement in accuracy and efficiency versus traditional computers, even assuming that non-cognitive computers could perform the tasks associated with the aspects disclosed herein. Therefore, the application of cognitive computers in particular allows for less calculations and less resource and power consumption than would otherwise be possible.

A First Exemplary Method of Providing a Custom Loan on a Per-User Basis

In one aspect, a computer-implemented method for determining a customized loan for a user may be provided. The method may include (1) receiving a request for a loan amount for a loan term; (2) calculating a statistical risk of default on the requested loan; (3) calculating a loan rate for the requested loan amount and loan term based upon the statistical risk of default; (4) receiving user input data including (i) demographic data for the user, and (ii) user behavioral data associated with the user; (5) calculating an adjusted statistical risk of default on the requested loan based upon the user input data; (6) adjusting the calculated loan rate for the requested loan to an adjusted loan rate based upon the adjusted statistical risk of default; and/or (7) presenting the adjusted loan rate to the user. The method may include additional, less, or alternate components, including those discussed elsewhere herein.
For instance, in various aspects, the method may include calculating the user's potential future earnings over the loan term based upon the user behavioral data, and calculating the adjusted statistical risk of default based upon the user's current earnings from the user input data and the user's potential future earnings.
Additionally or alternatively, the method may include calculating the user's potential future earnings by determining whether the user's is currently participating in one or more training programs or college courses that will increase the user's current earnings within the loan term. In such aspects, the user's potential future earnings may be calculated by identifying other user's having a demographic profile that matches that of the user, the other users having completed the same one or more training programs or college courses in which the user is currently participating, and calculating, as the user's potential future earnings, an average income of each of the other users at a future point within the loan term.
Furthermore, when the user's potential future earnings are used as part of this calculation, the method may include decreasing the calculated loan rate to a lower loan rate when the user's potential future earnings are greater than the user's current earnings in excess of a threshold amount. This decrease may include, for example, decreasing the calculated loan rate to the lower loan rate by an amount that is proportional to an amount in which the user's potential future earnings exceed the user's current earnings
Still further, the specifics of the loan may be sent to the user as a notification and/or viewed by the user when responding to the notification, such as via the user's computing device, for example. Thus, in various aspects, the method may include transmitting a notification to a computing device associated with the user via wireless communication or data transmission over one or more radio links or wireless communication channels, and presenting the adjusted loan rate to the user by displaying the adjusted loan rate on a display associated with the computing device in response to the user responding to the notification.

A Second Exemplary Method of Providing a Custom Loan on a Per-User Basis

In another aspect, a computer-implemented method for determining a customized loan for a user may be provided. The method may include (1) receiving a requested loan amount for a loan term, such as from user mobile or computing device via wireless communication or data transmission over one or more radio links or wireless communication channels; (2) calculating a statistical risk of default on the requested loan; (3) calculating a loan rate for the requested loan amount and loan term based upon the statistical risk of default; (4) receiving user input data including (i) demographic data for the user, and (ii) user behavioral data; (5) calculating an adjusted statistical risk of default on the requested loan based upon the user input data; (6) adjusting the calculated loan rate for the requested loan to an adjusted loan rate based upon the adjusted statistical risk of default; and/or (7) presenting the adjusted loan rate to the user, such as by transmitting the adjusted loan rate in an electronic message to the user's mobile or computing device via wireless communication or data transmission over one or more radio links or wireless communication channels. The method may include additional, less, or alternate components, including those discussed elsewhere herein.
For instance, in various aspects, the method may include calculating the user's potential future earnings over the loan term based upon the user behavioral data, and calculating the adjusted statistical risk of default based upon the user's current earnings from the user input data and the user's potential future earnings.
Additionally or alternatively, the method may include calculating the user's potential future earnings by determining whether the user's is currently participating in one or more training programs or college courses that will increase the user's current earnings within the loan term. In such aspects, the user's potential future earnings may be calculated by identifying other user's having a demographic profile that matches that of the user, the other users having completed the same one or more training programs or college courses in which the user is currently participating, and calculating, as the user's potential future earnings, an average income of each of the other users at a future point within the loan term.
Furthermore, when the user's potential future earnings are used as part of this calculation, the method may include decreasing the calculated loan rate to a lower loan rate when the user's potential future earnings are greater than the user's current earnings in excess of a threshold amount. This decrease may include, for example, decreasing the calculated loan rate to the lower loan rate by an amount that is proportional to an amount in which the user's potential future earnings exceed the user's current earnings
Still further, the specifics of the loan may be sent to the user as a notification and/or viewed by the user when responding to the notification, such as via the user's computing device, for example. Thus, in various aspects, the method may include transmitting a notification to a computing device associated with the user via wireless communication or data transmission over one or more radio links or wireless communication channels, and presenting the adjusted loan rate to the user by displaying the adjusted loan rate on a display associated with the computing device in response to the user responding to the notification.

Exemplary System for Providing a Custom Loan on a Per-User Basis

In yet another aspect, a computer system for predicting when a user requires a loan and presenting the user with a loan offer may be provided. The system may include (1) a client device (or mobile device) associated with a user, which may be configured to transmit a request for a loan amount for a loan term via wireless communication or data transmission over one or more radio links or wireless communication channels; and (2) one or more back-end components configured to (a) receive the request for the loan amount for the loan term; (b) calculate a statistical risk of default on the requested loan; (c) calculate a loan rate for the requested loan amount and loan term based upon the statistical risk of default; (d) receive user input data including (i) demographic data for the user, and (ii) user behavioral data; (e) calculate an adjusted statistical risk of default on the requested loan based upon the user input data; (f) adjust the calculated loan rate for the requested loan to an adjusted loan rate based upon the adjusted statistical risk of default; and/or (g) transmit a notification including the adjusted loan rate to the client device via wireless communication or data transmission over one or more radio links or wireless communication channels. The system may include additional, less, or alternate components, including those discussed elsewhere herein.
For instance, in various aspects, the one or more back-end components may be further configured to calculate the user's potential future earnings over the loan term based upon the user behavioral data, and to calculate the adjusted statistical risk of default based upon the user's current earnings from the user input data and the user's potential future earnings.
Additionally or alternatively, the one or more back-end components may be further configured to calculate the user's potential future earnings by determining whether the user is currently participating in one or more training programs or college courses that will increase the user's current earnings within the loan term. In such aspects, the user's potential future earnings may be calculated by identifying other user's having a demographic profile that matches that of the user, the other users having completed the same one or more training programs or college courses in which the user is currently participating, and calculating, as the user's potential future earnings, an average income of each of the other users at a future point within the loan term.
Furthermore, when the user's potential future earnings are used as part of this calculation, the one or more back-end components may be configured to decrease the calculated loan rate to a lower loan rate when the user's potential future earnings are greater than the user's current earnings in excess of a threshold amount. This may include, for example, decreasing the calculated loan rate to the lower loan rate by an amount that is proportional to an amount in which the user's potential future earnings exceed the user's current earnings.
Still further, the specifics of the loan may be sent to the client device as a notification and/or viewed by the user when responding to the notification via the client device, for example. Thus, in various aspects, the one or more back-end components may be configured to transmit a notification to the client device via wireless communication or data transmission over one or more radio links or wireless communication channels, and the client device may present the adjusted loan rate to the user via a suitable display in response to the user responding to the notification.

Cognitive Computing & Machine Learning

The cognitive computing and/or predictive modeling techniques discussed herein may include machine learning techniques or algorithms. For instance, customer data may be input into machine learning programs may be trained to (i) determine a statistical likelihood that a customer is looking for a loan, or may default on a loan, (ii) customize a loan product, and/or (iii) predict a life event, based upon the customer data, such as the types of customer data discussed elsewhere herein.
In certain embodiments, the cognitive computing and/or predictive modeling techniques discussed herein may heuristic engine and algorithms, and/or machine learning, cognitive learning, deep learning, combined learning, and/or pattern recognition techniques. For instance, a processor or a processing element may be trained using supervised or unsupervised machine learning, and the machine learning program may employ a neural network, which may be a convolutional neural network, a deep learning neural network, or a combined learning module or program that learns in two or more fields or areas of interest. Machine learning may involve identifying and recognizing patterns in existing data in order to facilitate making predictions for subsequent data. Models may be created based upon example inputs in order to make valid and reliable predictions for novel inputs.
Additionally or alternatively, the machine learning programs may be trained by inputting sample data sets or certain data into the programs, such as de-personalized customer data, image, mobile device, insurer database, and/or third-party database data. The machine learning programs may utilize deep learning algorithms that may be primarily focused on pattern recognition, and may be trained after processing multiple examples. The machine learning programs may include Bayesian program learning (BPL), voice recognition and synthesis, image or object recognition, optical character recognition, and/or natural language processing—either individually or in combination. The machine learning programs may also include natural language processing, semantic analysis, automatic reasoning, and/or machine learning.
In supervised machine learning, a processing element may be provided with example inputs and their associated outputs, and may seek to discover a general rule that maps inputs to outputs, so that when subsequent novel inputs are provided the processing element may, based upon the discovered rule, accurately predict the correct output. In unsupervised machine learning, the processing element may be required to find its own structure in unlabeled example inputs. In one embodiment, machine learning techniques may be used to identify potential loan applicants and customize loan products for individual customers.
In one embodiment, a processing element (and/or heuristic engine or algorithm discussed herein) may be trained by providing it with a large sample of images and/or user data with known characteristics or features. Based upon these analyses, the processing element may learn how to identify characteristics and patterns that may then be applied to analyzing user device details, user request or login details, user device sensors, geolocation information, image data, the insurer database, a third-party database, and/or other data. For example, the processing element may learn, with the user's permission or affirmative consent, to identify the user and/or the asset that is to be the subject of a transaction, such as generating an insurance quote or claim, opening a financial account, handling a loan or credit application, processing a financial (such as a credit card) transaction or the like.

Additional Considerations

This detailed description is to be construed as exemplary only and does not describe every possible embodiment, as describing every possible embodiment would be impractical, if not impossible. One may be implement numerous alternate embodiments, using either current technology or technology developed after the filing date of this application.
Further to this point, although the embodiments described herein often utilize credit report information as an example of sensitive information, the embodiments described herein are not limited to such examples. Instead, the embodiments described herein may be implemented in any suitable environment in which it is desirable to identify and control specific type of information. For example, the aforementioned embodiments may be implemented by a financial institution to identify and contain bank account statements, brokerage account statements, tax documents, etc. To provide another example, the aforementioned embodiments may be implemented by a lender to not only identify, re-route, and quarantine credit report information, but to apply similar techniques to prevent the dissemination of loan application documents that are preferably delivered to a client for signature in accordance with a more secure means (e.g., via a secure login to a web server) than via email.
Furthermore, although the present disclosure sets forth a detailed description of numerous different embodiments, it should be understood that the legal scope of the description is defined by the words of the claims set forth at the end of this patent and equivalents. The detailed description is to be construed as exemplary only and does not describe every possible embodiment since describing every possible embodiment would be impractical. Numerous alternative embodiments may be implemented, using either current technology or technology developed after the filing date of this patent, which would still fall within the scope of the claims. Although the following text sets forth a detailed description of numerous different embodiments, it should be understood that the legal scope of the description is defined by the words of the claims set forth at the end of this patent and equivalents. The detailed description is to be construed as exemplary only and does not describe every possible embodiment since describing every possible embodiment would be impractical. Numerous alternative embodiments may be implemented, using either current technology or technology developed after the filing date of this patent, which would still fall within the scope of the claims.
The following additional considerations apply to the foregoing discussion. Throughout this specification, plural instances may implement components, operations, or structures described as a single instance. Although individual operations of one or more methods are illustrated and described as separate operations, one or more of the individual operations may be performed concurrently, and nothing requires that the operations be performed in the order illustrated. Structures and functionality presented as separate components in example configurations may be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of the subject matter herein.
Additionally, certain embodiments are described herein as including logic or a number of routines, subroutines, applications, or instructions. These may constitute either software (e.g., code embodied on a machine-readable medium or in a transmission signal) or hardware. In hardware, the routines, etc., are tangible units capable of performing certain operations and may be configured or arranged in a certain manner. In example embodiments, one or more computer systems (e.g., a standalone, client or server computer system) or one or more hardware modules of a computer system (e.g., a processor or a group of processors) may be configured by software (e.g., an application or application portion) as a hardware module that operates to perform certain operations as described herein.
In various embodiments, a hardware module may be implemented mechanically or electronically. For example, a hardware module may comprise dedicated circuitry or logic that is permanently configured (e.g., as a special-purpose processor, such as a field programmable gate array (FPGA) or an application-specific integrated circuit (ASIC)) to perform certain operations. A hardware module may also comprise programmable logic or circuitry (e.g., as encompassed within a general-purpose processor or other programmable processor) that is temporarily configured by software to perform certain operations. It will be appreciated that the decision to implement a hardware module mechanically, in dedicated and permanently configured circuitry, or in temporarily configured circuitry (e.g., configured by software) may be driven by cost and time considerations.
Accordingly, the term “hardware module” should be understood to encompass a tangible entity, be that an entity that is physically constructed, permanently configured (e.g., hardwired), or temporarily configured (e.g., programmed) to operate in a certain manner or to perform certain operations described herein. Considering embodiments in which hardware modules are temporarily configured (e.g., programmed), each of the hardware modules need not be configured or instantiated at any one instance in time. For example, where the hardware modules comprise a general-purpose processor configured using software, the general-purpose processor may be configured as respective different hardware modules at different times. Software may accordingly configure a processor, for example, to constitute a particular hardware module at one instance of time and to constitute a different hardware module at a different instance of time.
Hardware modules may provide information to, and receive information from, other hardware modules. Accordingly, the described hardware modules may be regarded as being communicatively coupled. Where multiple of such hardware modules exist contemporaneously, communications may be achieved through signal transmission (e.g., over appropriate circuits and buses) that connect the hardware modules. In embodiments in which multiple hardware modules are configured or instantiated at different times, communications between such hardware modules may be achieved, for example, through the storage and retrieval of information in memory structures to which the multiple hardware modules have access. For example, one hardware module may perform an operation and store the output of that operation in a memory device to which it is communicatively coupled. A further hardware module may then, at a later time, access the memory device to retrieve and process the stored output. Hardware modules may also initiate communications with input or output devices, and may operate on a resource (e.g., a collection of information).
The various operations of example methods described herein may be performed, at least partially, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors may constitute processor-implemented modules that operate to perform one or more operations or functions. The modules referred to herein may, in some example embodiments, comprise processor-implemented modules.
Similarly, the methods or routines described herein may be at least partially processor-implemented. For example, at least some of the operations of a method may be performed by one or more processors or processor-implemented hardware modules. The performance of certain of the operations may be distributed among the one or more processors, not only residing within a single machine, but deployed across a number of machines. In some example embodiments, the processor or processors may be located in a single location (e.g., within a home environment, an office environment or as a server farm), while in other embodiments the processors may be distributed across a number of locations.
The performance of certain of the operations may be distributed among the one or more processors, not only residing within a single machine, but deployed across a number of machines. In some example embodiments, the one or more processors or processor-implemented modules may be located in a single geographic location (e.g., within a home environment, an office environment, or a server farm). In other example embodiments, the one or more processors or processor-implemented modules may be distributed across a number of geographic locations.
Unless specifically stated otherwise, discussions herein using words such as “processing,” “computing,” “calculating,” “determining,” “presenting,” “displaying,” or the like may refer to actions or processes of a machine (e.g., a computer) that manipulates or transforms data represented as physical (e.g., electronic, magnetic, or optical) quantities within one or more memories (e.g., volatile memory, non-volatile memory, or a combination thereof), registers, or other machine components that receive, store, transmit, or display information.
As used herein any reference to “one embodiment” or “an embodiment” means that a particular element, feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.
Some embodiments may be described using the expression “coupled” and “connected” along with their derivatives. For example, some embodiments may be described using the term “coupled” to indicate that two or more elements are in direct physical or electrical contact. The term “coupled,” however, may also mean that two or more elements are not in direct contact with each other, but yet still co-operate or interact with each other. The embodiments are not limited in this context.
As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Further, unless expressly stated to the contrary, “or” refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by any one of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).
In addition, use of the “a” or “an” are employed to describe elements and components of the embodiments herein. This is done merely for convenience and to give a general sense of the description. This description, and the claims that follow, should be read to include one or at least one and the singular also includes the plural unless it is obvious that it is meant otherwise.
The patent claims at the end of this patent application are not intended to be construed under 35 U.S.C. § 112(f) unless traditional means-plus-function language is expressly recited, such as “means for” or “step for” language being explicitly recited in the claim(s).
The various systems and methods described herein are directed to an improvement to computer functionality, and improve the functioning of conventional computers, as described, for example, in the “Technical Advantages” Section and elsewhere herein.

Claims

1. A computer-implemented method, comprising:

training, by one or more processors, a machine learning model to predict a set of historical earnings values associated with a set of users, wherein the training comprises determining at least one correlation between a set of factors within demographic profiles of the set of users and the set of historical earnings values;

receiving, by the one or more processors, a request for a loan for a user, the request indicating a loan amount and a loan term;

determining, by the one or more processors, an initial statistical risk of default on the loan, based at least in part on user profile data associated with the user that indicates a current earnings value of the user;

determining, by the one or more processors, an initial loan rate for the loan, based at least in part upon the initial statistical risk of default;

generating, by the one or more processors and using the machine learning model, a predicted future earnings value associated with the user over the loan term, based at least in part on whether the user profile data indicates that the user is currently enrolled in an educational program that the machine learning model predicts will cause an increase in earnings of the user, during the loan term, relative to the current earnings value of the user;

generating, by the one or more processors, an adjusted statistical risk of default on the loan by adjusting the initial statistical risk of default, based at least in part upon a difference between the current earnings value and the predicted future earnings value;

generating, by the one or more processors, an adjusted loan rate for the loan by adjusting the initial loan rate, based at least in part upon the adjusted statistical risk of default; and

causing, by the one or more processors, a computing device associated with the user to present the adjusted loan rate.

2. (canceled)

3. (canceled)

4. The computer-implemented method of claim 1, wherein generating the predicted future earnings value comprises:

identifying, by the one or more processors, a subset of the demographic profiles that is associated with other users who have completed the same educational program in which the user is currently enrolled; and

determining, by the one or more processors, the predicted future earnings value associated with the user based at least in part on an average of the historical earnings values associated with the other users.

5. The computer-implemented method of claim 1, wherein generating the adjusted loan rate comprises:

determining, by the one or more processors, that the predicted future earnings value is greater than the current earnings value and the difference between the current earnings value and the predicted future earnings value exceeds a threshold amount; and

generating, by the one or more processors, the adjusted loan rate by decreasing the initial loan rate.

6. The computer-implemented method of claim 5, wherein generating the adjusted loan rate by decreasing the initial loan rate comprises:

decreasing, by the one or more processors, the initial loan rate by an amount that is proportional to the difference between the current earnings value and the predicted future earnings value.

7. The computer-implemented method of claim 1, wherein causing the computing device to present the adjusted loan rate comprises:

causing, by the one or more processors, transmission of a notification to the computing device via wireless communication or data transmission over one or more radio links or wireless communication channels, and

wherein the notification causes a display associated with the computing device to present the adjusted loan rate.

8. A computer system, comprising:

one or more processors;

memory storing computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

training a machine learning model to predict a set of historical earnings values associated with a set of users, wherein the training comprises determining at least one correlation between a set of factors within demographic profiles of the set of users and the set of historical earnings values;

receiving a request for a loan for a user, the request indicating a loan amount and a loan term;

determining an initial statistical risk of default on the loan, based at least in part on user profile data associated with the user that indicates a current earnings value of the user;

determining an initial loan rate for the loan, based at least in part upon the initial statistical risk of default;

generating, using the machine learning model, a predicted future earnings value associated with the user over the loan term, based at least in part on whether the user profile data indicates that the user is currently enrolled in an educational program that the machine learning model predicts will cause an increase in earnings of the user, during the loan term, relative to the current earnings value of the user;

generating an adjusted statistical risk of default on the loan by adjusting the initial statistical risk of default, based at least in part upon a difference between the current earnings value and the predicted future earnings value;

generating an adjusted loan rate for the loan by adjusting the initial loan rate, based at least in part upon the adjusted statistical risk of default; and

causing a client device associated with the user to present the adjusted loan rate.

9. (canceled)

10. (canceled)

11. The computer system of claim 8, wherein generating the predicted future earnings value comprises:

identifying a subset of the demographic profiles that is associated with other users who have completed the same educational program in which the user is currently enrolled, and

determining the predicted future earnings value associated with the user based at least in part on an average of the historical earnings values associated with the other users.

12. The computer system of claim 8, wherein generating the adjusted loan rate comprises:

determining that the predicted future earnings value is greater than the current earnings value and the difference between the current earnings value and the predicted future earnings value exceeds a threshold amount; and

generating the adjusted loan rate by decreasing the initial loan rate.

13. The computer system of claim 12, wherein generating the adjusted loan rate by decreasing the initial loan rate comprises:

decreasing the initial loan rate by an amount that is proportional to the difference between the current earnings value and the predicted future earnings value.

14. The computer system of claim 8, wherein causing the client device to present the adjusted loan rate comprises:

causing transmission of a notification to the client device via wireless communication or data transmission over one or more radio links or wireless communication channels, and

wherein the notification causes a display of the client device to present the adjusted loan rate.

15. A non-transitory, tangible computer-readable medium storing instructions thereon, that, when executed by one or more processors, cause the one or more processors to:

train a machine learning model to predict a set of historical earnings values associated with a set of users, wherein training the machine learning model comprises determining at least one correlation between a set of factors within demographic profiles of the set of users and the set of historical earnings values;

receive a request for a loan for a user, the request indicating a loan amount and a loan term;

determine an initial statistical risk of default on the loan, based at least in part on user profile data associated with the user that indicates a current earnings value of the user;

determine an initial loan rate for the loan, based at least in part upon the initial statistical risk of default;

generating an adjusted statistical risk of default on the loan, based at least in part upon a difference between the current earnings value and the predicted future earnings value;

16. (canceled)

17. (canceled)

18. The non-transitory, tangible, computer-readable medium of claim 15, wherein the instructions cause the one or more processors to generate the predicted future earnings value by:

identifying a subset of the demographic profiles that is associated with other users who have completed the same educational program in which the user is currently enrolled; and

19. The non-transitory, tangible, computer-readable medium of claim 15, wherein the instructions cause the one or more processors to generate the adjusted loan rate by:

generating the adjusted loan rate by decreasing the initial loan rate.

20. The non-transitory, tangible, computer-readable medium of claim 19, wherein generating the adjusted loan rate by decreasing the initial loan rate comprises:

21. The computer-implemented method of claim 1, wherein the computing device is a mobile device of the user, and the one or more processors receive the request for the loan from the mobile device.

22. The computer-implemented method of claim 1, wherein generating the predicted future earnings value is further based at least in part on one or more predicted life events associated with the user, the one or more predicted life events being predicted by the machine learning model based at least in part on the user profile data.

23. The computer-implemented method of claim 22, wherein the one or more predicted life events includes at least one of: the user getting married, a child associated with the user attending college, the user paying off a previous loan, the user receiving a settlement, or the user receiving an inheritance.

24. The computer system of claim 8, wherein generating the predicted future earnings value is further based at least in part on one or more predicted life events associated with the user, the one or more predicted life events being predicted by the machine learning model based at least in part on the user profile data, the one or more predicted life events including at least one of: the user getting married, a child associated with the user attending college, the user paying off a previous loan, the user receiving a settlement, or the user receiving an inheritance.

25. (canceled)

26. The non-transitory, tangible, computer-readable medium of claim 15, wherein the instructions cause the one or more processors to generate the predicted future earnings value further based at least in part on one or more predicted life events associated with the user, the one or more predicted life events being predicted by the machine learning model based at least in part on the user profile data, the one or more predicted life events including at least one of: the user getting married, a child associated with the user attending college, the user paying off a previous loan, the user receiving a settlement, or the user receiving an inheritance.

27. The computer-implemented method of claim 1, wherein the set of factors within the demographic profiles of the set of users includes educational programs associated with the set of users.