System and method for providing an aggregation tool

Abstract

Embodiments of the present invention assist in the development, management, and deployment of aggregated data attributes for multiple data sources. One embodiment provides a development interface that allows for elements of attributes, including filters, to be moved into a coding area in which an attribute or an attribute element is being edited. In another embodiment, the user interface presents data fields to assist in the development of filters for multiple data sources with divergent formats. The application further provides a validation interface through which users can validate attributes and trace the results returned by various elements referenced by the attributes under validation. Another embodiment provides a system for managing attributes and deploying them to various systems by creating a deployment file that is used by an attribute calculation system. In one embodiment, the attribute calculation system is a scalable system that dynamically calculates attributes for multiple data sources.

Description

BACKGROUND OF THE INVENTION

Field of the Invention

The disclosure relates generally to the field of scoring and prediction and to tools to improve the flexibility and efficiency in generating and evaluating aggregated attributes.

Description of the Related Art

There exists significant interest in information indicative of the relative financial risk or profitability of potential business transactions with individuals or other entities. For example, a lending institution is interested in the relative likelihood of a loan recipient timely and reliably making mutually agreed loan payments. An insurance company is interested in the relative likelihood of claims arising from an existing or potential client. Such predictive information is a factor in deciding whether to engage in a particular business transaction and/or the terms under which a transaction takes place.

A large variety of public records and privately developed databases can be utilized to inform such risk/benefit determinations. For example, credit reporting agencies (CRAs) collect and maintain information on a person's individual credit history. This information can include a total credit line on one or more accounts, current credit balance, credit ratios, satisfactorily paid accounts, any late payments or delinquencies, depth of credit history, total outstanding credit balance, and/or records of recent and/or historical inquiries into the person's credit. Governmental motor vehicle agencies generally maintain records of any vehicle code violations by a person as well as histories of reported accidents. Courts will generally maintain records of pending or disposed cases associated with a person, such as small claims filings, bankruptcy filings, and/or any criminal charges. Similar information also exists for large and small businesses, such as length of the business's existence, reported income, profits, outstanding accounts receivable, payment history, market share, and so forth.

The extensive amount of data available for any given person or entity makes the task of evaluating a business decision regarding the person or entity very difficult. Accordingly, such raw data is frequently processed to facilitate more convenient and rapid financial decisions. For example, a person's raw financial data can be processed to produce a “score” indicative of their relative credit worthiness. Such a score can be utilized in decisions to extend the person or entity further credit and/or as a factor in determining an interest rate to be charged. The evaluation of the relative risk/benefit of a given decision is even more complex when considering multiple persons simultaneously, such as spouses, partnerships, sole proprietorships, joint ventures, LLCs or other entities. When considering multiple persons, raw data from multiple sources about each of the individuals may need to be evaluated.

Attributes can be used to calculate various types of scores and in many instances may be used on their own to guide business decisions as well. Attributes can be aggregated to target various aspects of credit histories, bankruptcy data, and other types of non-credit-based data. For example, a simple attribute could be “consumers who have opened a new credit line in the last 12 months.” The results of the attribute would be a set of consumers who meets both the criteria of having opened a new credit line and having done so in the last 12 months. Therefore, attributes are important in facilitating the use of raw data for a variety of decisions that financial institutions and other entities may need to make.

SUMMARY OF THE DISCLOSURE

Embodiments of the present invention provide systems and methods for creating, managing, and calculating attributes and filters for multiple data sources. One embodiment is an attribute toolbox (ATB) architecture that comprises an ATB user interface (ATB UI) and an ATB engine system.

One embodiment of the present invention is a system for creating and using credit data attributes, comprising an attribute repository for storing a plurality of credit data attributes, a first application server executing a platform-independent user interface application, a credit data repository, and a second application server executing an attribute calculation engine system, wherein the attribute calculation engine system is configured to retrieve credit data from the credit data repository and performs calculations on the credit data in accordance to the attributes in a deployment package generated by the user interface application. In one embodiment, the user interface application further comprises a development interface for editing credit data attributes, wherein the development interface is configured to retrieve the credit data attributes from the attribute repository for editing and return edited credit data attributes or newly created credit data attributes for saving in the attribute repository, a validation interface for testing credit data attributes, wherein the user interface application is configured to execute a validation engine to calculate credit data attributes on test credit data and display the results on the validation interface, and a management interface for managing the deployment of the credit data attributes, wherein the credit data attributes are saved into a deployment package.

Another embodiment is a system for creating of data attributes, comprising an attribute repository and a user interface application configured to receive a plurality of requests from users. The attribute repository stores a plurality of data attribute elements, including attributes, filters, functions, operators, and data definitions. The user interface application further comprises a development interface for editing attribute elements, wherein the development interface is configured to retrieve attribute elements from the attribute repository for editing and return edited attribute elements or newly created attribute elements for saving in the attribute repository, and a validation interface for testing data attributes, wherein the user interface application is configured to execute a validation processing engine to calculate data attributes on test credit data and display the calculation results of the individual attribute elements on the validation interface.

One embodiment is a system for calculating credit data attributes, comprising at least one credit data repository, a first plurality of application servers executing a plurality of attribute calculation engines, a second plurality of application servers executing a plurality of data access engines, wherein the data access engines are configured to access data from the at least one credit data repository, and a third plurality of application servers executing a plurality of façade applications, wherein the façade applications are configured to receive a plurality of attribute calculation requests from a plurality of calling applications, send corresponding requests to the attribute calculation engines and data access engines, and return calculation results from the attribute calculation engines to the calling applications.

One embodiment is a method for building attributes for data from multiple data sources, comprising the steps of defining a plurality of filters, wherein each filter takes into account the format of a data source among a plurality of data sources, defining a plurality of data source-independent attributes, wherein the attributes operate on data selected by the plurality of filters, and creating a deployment package based on the plurality of attributes.

One embodiment is a method for calculating data scores based on data from multiple data sources, comprising the steps of selecting a plurality of data sets from a plurality of data sources, wherein each of the plurality of data sources stores its data in a different format, defining a plurality sets of filters, wherein each set of filters takes into account the format of each of the plurality of data sources, defining a plurality of data source-independent attributes, extracting a subset of data from each of the selected data sets in accordance to the filters, calculating the attributes based on the subset of data, and aggregating the results obtained from the step of calculating to create data scores.

One embodiment of the present invention is a system for creating and using data attributes, comprising an attribute repository for storing a plurality of data attributes, a first application server executing a platform-independent user interface application, a data repository, and a second application server executing an attribute calculation engine system, wherein the attribute calculation engine system is configured to retrieve data from the data repository and performs calculations on the data in accordance to the attributes in a deployment package generated by the user interface application. In one embodiment, the user interface application further comprises a development interface for editing data attributes, wherein the development interface is configured to retrieve the data attributes from the attribute repository for editing and return edited data attributes or newly created data attributes for saving in the attribute repository, a validation interface for testing data attributes, wherein the user interface application is configured to execute a validation engine to calculate data attributes on test data and display the results on the validation interface, and a management interface for managing the deployment of the data attributes, wherein the data attributes are saved into a deployment package.

BRIEF DESCRIPTION OF THE DRAWINGS

Specific embodiments of the invention will now be described with reference to the following drawings, which are intended to illustrate embodiments of the invention, but not limit the invention:

FIG. 1 illustrates the architecture of the ATB architecture according to an embodiment;

FIG. 2 illustrates the hierarchy and the dependencies of systems, clients, attributes, filters, functions and raw data as used in one embodiment;

FIG. 3A illustrates an example deployment configuration of the ATB UI according to an embodiment;

FIG. 3B illustrates an example deployment configuration of the ATB engine system according to an embodiment;

FIG. 4A is a schematic diagram showing the implementation of the ATB UI according to one embodiment;

FIG. 4B is a schematic diagram showing the implementation of the ATB engine system according to one embodiment;

FIG. 5A is an example computer screen showing how attributes are located and managed in the ATB UI according to one embodiment;

FIG. 5B is an example computer screen showing the attribute development interface according to one embodiment;

FIG. 5C is an example computer screen showing how functions and operators can be added to attributes according to one embodiment;

FIG. 6A is an example computer screen showing the filter development interface according to one embodiment;

FIG. 6B is an example computer screen showing how details of a selected data source are displayed in the filter development interface according to one embodiment;

FIG. 6C is a computer screen showing a search and browse interface displaying how filters are used according to one embodiment;

FIG. 7 is an example computer screen of the validation interface showing the results of the validation engine according to one embodiment;

FIG. 8A is an example computer screen showing how a client system can be created in the ATB UI according to one embodiment;

FIG. 8B is an example computer screen showing how attributes can be associated with a client system for deployment according to one embodiment;

FIG. 8C is an example computer screen showing how multiple clients and systems are managed in the ATB UI according to one embodiment;

FIG. 9A is an example computer screen showing how multiple data sources are managed in the ATB UI according to one embodiment; and

FIG. 9B is an example computer screen showing show attributes and filters can be exported according to one embodiment.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Embodiments of the ATB UI system architecture solve many problems associated with developing, managing, and deploying attributes. First, many common aggregated attributes are often complex, referencing other attributes as well as filters that are data format dependent. The filters themselves have to be coded to access specific fields within a data source. For example, since each of the main CRAs in the United States maintains data in a different format, a separate filter is often used for each CRA. Therefore, the development of filters and attributes is often a tedious process and syntax errors frequently occur. Those skilled in the art will recognize that attributes can be developed for non-credit data sources, and the challenge of accounting for the individual formats of non-credit data sources exist as well. In one embodiment, the ATB UI includes a development interface that allows for filters and attributes to be moved via the UI, for example, into a coding area in which an attribute or a filter is being edited. In another embodiment, the ATB UI provides a filter development interface that presents data fields to the user. The interface simplifies the task of developing filters for multiple data sources that have different formats by presenting the data fields of the data sources in a graphical user interface. The fields can be moved into coding area as well, ensuring that proper data field references are included.

Second, the development of efficient and accurate attributes relies on extensive testing on raw data and constant revisions are made based on test results. Attributes often reference other attribute elements in a long chain of dependencies, including other attributes that may in turn reference numerous filters. For example, in developing an attribute that uses five other attributes, a developer may have to run numerous tests in order to arrive at the proper weight that should be accorded to each of the five attributes. In one embodiment, the ATB UI includes a validation interface that advantageously allows the user to see results returned by each attribute element that is referenced by the attribute that is currently being tested. One embodiment displays the results of each filter and each attribute that is referenced by the attribute that is being validated. This enables the user to quickly determining any errors or performance problems hidden in the chain of dependencies. The user can, for example, quickly determine that a filter is selecting a wrong field or that an attribute calculation is returning erroneous or undesirable results. The interface is particularly helpful in the development of aggregation attributes, which often reference a large number attributes and filters. The ability to trace the results of the individual elements is invaluable to the development of such attributes.

Third, institutions often require the use hundreds of attributes in multiple combinations to handle complex analysis and modeling on credit-based and non-credit based data. As a result, both the creation and maintenance of filters and attributes can require significant time and effort. Finally, the ability to effectively calculate attributes and scores on data from multiple data sources also presents significant challenges as well.

Embodiments of the ATB UI and ATB engine system architectures facilitate the management of the attributes and remove some of the difficulties that come with having to manually configure the deployment of attributes in system files. In one embodiment, the ATB UI allows users to associate groups of attributes into a system through a graphical user interface. The ATB UI also provides a comprehensive view of how attributes are deployed in various systems. Users can quickly and easily view the dependencies between attributes and the associated systems and determine how editing an attribute will affect the systems in their deployment scheme. In one embodiment, upon user request the ATB UI will automatically output a single deployment package including all of the user's systems, along with the proper associations of attributes, filters, and other configuration information. The deployment package can then be used in the ATB engine system, a system that performs attribute calculations.

In one embodiment, the ATB engine system is configured to dynamically access multiple data sources and automatically performs attribute calculations based on the attributes included in the deployment package. In another embodiment, the ATB engine system can calculate the attributes on credit data for a related set of individuals or entities. In one embodiment, both the ATB UI and the ATB engine system are implemented in a component-based architecture with a high degree of flexibility, maintainability, and scalability. In one embodiment, multiple components can be hosted in a single environment. In another embodiment, the components can be distributed across a network with a load balancing server to manage and distribute requests among the components.

FIG. 1 depicts the architecture of an embodiment of the present invention. Attribute Toolbox (ATB) architecture 100 includes two main components: an ATB User Interface (ATB UI) module 102 and an ATB engine system 104. The ATB UI 102 presents a comprehensive user interface through which users can view, edit, validate, and simulate filters and attributes. The ATB engine system 104 takes the attributes created by the ATB UI 102, fetches data from the appropriate data sources, and performs calculations on the fetched data based on those attributes.

ATB UI

As shown in FIG. 1, the ATB UI 102 includes a presentation component 110, a business logic abstraction 112, an attribute/filter management business logic 114, and a data access layer 116.

Attribute Development

The presentation component 110 presents various user interfaces to assist in the development of filters and attributes and the management of their deployment to various systems. In one embodiment, the presentation component 110 further comprises filter and attribute development interfaces 150, a search/browse interface 152, a management interface 154, and a validation interface 156. The filter and attribute development interfaces 150 enable users to drag and drop existing data definitions, filters, functions, and attributes into a coding area and thereby substantially reducing the time and effort required to locate the proper elements that need to be added into new filters and attributes. The search interface 152 enables user to search for and browse filters, attributes, and clients/systems. The management interface 154 allows users to configure clients and systems and provides an interface through which attributes can be associated to systems for deployment. These user interfaces will be further described in conjunction with FIGS. 5A-9B. In one embodiment, the ATB UI 102 accepts requests from client computers 122 and responds by presenting the appropriate user interfaces to the client computers 122.

These user interfaces access objects and methods within the business logic abstraction 112 layer, which is an abstraction layer of the attribute/filter management business logic 114. Hence, any future changes in the implementation in the attribute/filter management business logic 114 will not necessarily necessitate recoding or reconfiguring the presentation component 110. The ATB UI 102 retrieves attributes and filters from the attribute/filter repository 118 through the data access layer 116. Likewise, new or edited filters and attributes are saved back to the repository 118. The attributes can be standardized attributes such as standard aggregation (STAGG) attributes created by Experian, custom-made attributes, or both.

In one embodiment, the repository 118 implements a version control feature so that users of the ATB UI 102 can retrieve previous versions of attributes and filters. Furthermore, the ATB UI 102 provides a locking mechanism by which users can lock the attributes and filters that they are currently editing to prevent other users from changing them. The locking mechanism and the version control feature enable multiple users to collaborate on the development of attributes and filters.

In addition, one embodiment of the ATB UI 102 has user access control. For example, some users may be given rights to execute but not view attributes. Other users may have full rights to view and edit the attributes. In one embodiment, the user access rights are controlled by special serial numbers or passcodes, though it is recognized that a variety of security procedures and methods may be used. A user presents the proper serial number or passcode in order to view and edit attributes. The modularity of various interfaces, in addition to the user control feature of ATB UI 102, allows users with various roles to collaborate on a project. For example, developers of attributes and filters may be granted full rights to view and modify attributes and filters, while business analysts may be granted limited rights to deploy the attributes to various systems. In another example, a company that specializes in developing attributes and filters may have full access rights while its customers who purchase the attributes and filters may be given limited rights to use but not view the attributes and filters.

Testing, Validation and Simulation

One main difficulty with attribute and filter development is the significant effort required for validation. Often attribute and filter developers have to test the attributes and filters on sample data and modify them repeatedly to improve performance and accuracy. In addition to this iterative process, developers have to account for the large number of dependencies that can potentially be in any attributes and filters. FIG. 2 illustrates the potential dependencies as used in one embodiment. Hierarchy 170 shows that clients are dependent on systems, which in turn are dependent on attributes. An example client may be a credit card company, and the company may have a number of systems for its credit card products. For example, one system may be used to screen commercial card applicants and another may be used to screen consumer card applicants. As such, the two systems have different attributes because the screening criteria for these two groups of applicants are different.

Returning to the hierarchy, attributes are further dependent on filters, which in turn are dependent on the raw data definitions. This is because attributes are data-independent and filters are data dependent. In one embodiment, a filter must be defined for each data source. As shown in the hierarchy 170, each filter can also reference other filters, functions and data definitions. Each attribute, in turn, can reference a number of filters, functions and other attributes. Thus, an error in a data definition in a single filter can adversely affect all filters that reference that single filter, and in turn all attributes that reference those affected filters will be affected as well, and so on.

One embodiment of the ATB UI 102 addresses this difficulty by providing a validation engine 160 that allows for dependency tracing. In one example, an attribute A includes three other attributes B, C, D, and attributes B, C and D each additionally has three filters B1, B2, B3 and so on. When the user decides to test attribute A on a test data set in the validation engine 160, the validation interface 156 shows the user the results of applying filters B1, B2, B3, C1, C2, C3, D1, D2, and D3 to the sample data, as well as the results of calculating attributes B, C, and D. In addition, the validation interface 156 displays the raw data used in testing. Because each result in the chain of dependency for attribute A is displayed, the user can quickly locate any problem within the chain of dependency. For example, the user can see that attribute B is returning undesirable results, determine that filter B1 is incorrect, and adjust filter B1 to correct the problem. Because the dependencies can often be far more numerous and complex than this simple example, the validation interface 156 provides a powerful tracing tool for developers of attributes and filters for multiple data sources.

In addition to the validation engine 160, embodiments also provide a simulation configuration wherein the user can utilize the ATB UI 102 to simulate the performance of their filters and attributes to see how they affect attribute and score calculations. In one embodiment, the ATB UI 102 provides an interface by which a user may connect a custom application so the user may simulate the attributes and filters created in the ATB UI 102 in the custom application.

ATB Viewer

Embodiments of the invention also include an ATB Viewer component. The ATB Viewer is a user interface that allows users to “view” parsed raw data, calculated filter or attribute results. Attributes and filters can be created and generated and the results can be stored. The results can further be returned to a custom/calling application selected by the user. If the user wishes to view the results immediately, he/she can access the ATB Viewer to have a general formatted overview of the results for a specific test case.

Management and Deployment

The ATB UI 102 allows users to manage the collections of filters and attributes that are slated for deployment in the ATB engine system 104. Users can quickly create attributes and associate a group of attributes for deployment. For example, a bank user may create twenty attributes for its credit card operations and another fifty attributes for its home loan operations. The bank may then associate the twenty attributes to a “Credit Cards” system and the fifty attributes to a “Home Loans” systems by using the ATB UI 102. The ATB UI 102 then outputs a deployment package 120 to be used by the ATB UI engine system 104, which performs calculations based on these associated attributes included with the deployment package 120. In one embodiment, the deployment package 120 is generated with the proper dependencies among the attributes and filters. Furthermore, the deployment package 120 can accommodate multiple systems, each with its own associated attributes. This one-step deployment function of this embodiment substantially eliminates the cost of having to code the dependencies of the attributes manually and enables rapid deployment of multiple systems with a single request sent to the ATB UI 102.

ATB Engine System

The ATB engine system 104 retrieves and parses raw data, calculates attributes, and returns results. In the illustrated embodiment, the ATB engine system 104 comprises an ATB engine 106 and a data access engine 108. In other embodiments, they can be combined into one engine. The ATB engine 106 calculates attributes or scores using data received from the data access engine 108. In one embodiment, the ATB engine 106 comprises a parser 162, which reads and parses data. In other embodiments, during calculation, the ATB engine 106 reads instructions in a manually created file or a deployment file produced by the ATB UI 102. The ATB engine 106 then instructs a business rules component 164, which performs the actual calculations.

In one embodiment, the data access engine 108 accesses the external data source(s) 140 or reads the data from one or several files (for example, an automatic response format (ARF) file), parses the data, and sends it to the ATB engine 106 for calculation. The data access engine 108 comprises a mapper 166 and a communication module 168. The mapper 166 handles the mapping of data fields in data sources 140 to objects within ATB engine system 104. The communication module 168 handles the tasks of communicating with various data sources 140. Because data access engine 108 accepts a wide variety of data input, it can be configured to access data from credit data sources 402 as well as data from non-credit data sources 404, such as a local data source 406. In one embodiment, the local data source 406 is a batch XML input. For example, a bank running an instance of the ATB engine 104 may configure the ATB engine 104 to perform attribute calculations on data from a credit bureau, a third-party data source, and/or its own internal customer data. In one embodiment, the results of the calculations are sent to a repository 148 so the calling application 124 may access them.

Because the ATB engine system 104 can handle multiple data sources, it can dynamically calculate the attributes across data sources as needed. Embodiments of the ATB engine system 104 can be configured to automatically calculate attributes across multiple data sources, whether in response to requests from a plurality of calling applications or a large volume of batch requests from batch files. The ability to handle calculations across multiple data sources in accordance to the incoming requests greatly assist in the evaluation of the relative risk/benefit of a given financial decision that involve multiple persons at the same time, such as spouses, partnerships, sole proprietorships, joint ventures, LLCs or other entities. Thus, unlike prior systems, users of the ATB engine system 104 will not have to access the data sources individually and perform attribute calculation for each source and then combine the results of their calculations. Similarly, the ATB engine system 104 will not have to perform an analysis on individuals separately and then combine the results.

Profile

Embodiments of the ATB engine system 104 additionally includes a profile component. In one embodiment the profile component is part of the façade 130. A profile provides the ability to handle simple work flows. For example, one work flow may be a decision of call one data source versus another that depends on the results coming from a previous service request. For example, one profile can include instructions to the ATB engine system 104 to access an alternative data source in case a first request does not return the desired data. The profile workflow logic can include a variety of instructions and requests that can be executed based on the occurrence or non-occurrence of various conditions. Embodiments of ATB engine system include profiles that can handle roll-over logic, ZIP code tables and any other type of logic driven by one or several attribute calculations.

ATB UI Architecture

FIG. 3A depicts an embodiment of the ATB UI web server architecture. The illustrated ATB UI architecture 200 includes an ATB UI web server 202 configured to receive requests from a plurality of workstations or other remote computing devices 204. In one embodiment, workstations or remote computing devices 204 send requests and receive responses via the HTTP or HTTPS protocols, and use web browsers to solicit inputs and display the information to the end users. For example, an end user may be a bank that regularly extends credit to its customers. The bank may be able to use its computers to remotely interact with the ATB UI web server 202 in order to view, edit, validate, and test its own attributes and filters stored in relational databases 206. In one embodiment, the ATB UI web server 202 is a self-contained Ruby on Rails application that acts as a web server. It communicates via the HTTP or HTTPS protocol in one embodiment. Syntax checking of attribute codes is accomplished through java-script on the client computers. It is recognized that other applications, protocols, and languages may be used.

ATB Engine Architecture

FIG. 3B illustrates an embodiment of the ATB engine system architecture. The illustrated ATB engine system architecture 220 includes a calling application 222, a load balancer 224, one or more façade web services 226, data access engines 228, ATB engines 230, and a data storage 232. In one embodiment, the calling application 222 is an application that sends calculation requests to the load balancer 224. In other embodiments, the calling application 222 directly sends requests to the façade web service 226. These requests include requests to calculate attributes and scores. For example, a bank may have a calling application that sends a request to calculate the number of customers who have the attribute of recently opening a credit line within the last twelve months. In one embodiment, the calling application is a custom-coded application. For example, a mortgage company may have a custom-coded application to process its loan applicants. In another embodiment, the calling application is a commercially available application such as Transact by Experian Decision Analytics. It is recognized that other applications may be used.

In one embodiment, the load balancer 224 receives requests from a plurality of calling applications 222 and distributes those requests to a plurality of the façade web services 226 in order to improve the efficiency of the overall architecture. Each façade web service 226 publishes objects and methods available for attribute calculation. Upon receipt of an incoming request from the calling application 224, the façade web service 226 initiates the appropriate remote procedural calls (RPC) over the TCP/IP protocol to either the data access engine 228 or the ATB engine 230. For example, the façade web service 226 may initiate a first RPC to the data access engine 228 to retrieve the requested data from the data storage 232 and then initiate a second RPC to the ATB engine 230 for calculation. The ATB engine 230 then calculates the results by applying attributes within the deployment package 120 to the retrieved data sets. In one embodiment, a user can use the ATB UI 102 to create the deployment package 120. The final results are sent back to the calling application 222. In the recently opened credit line example mentioned above, the results would be a list of customers who have opened a recent credit line.

The configuration depicted in FIGS. 3A and 3B is for illustrative purposes only and those skilled in the art will recognize that any number of components of the architecture can be used. In one embodiment, the architecture is designed to be scalable. For example, although not depicted, a plurality of calling applications and a plurality of credit bureau data sources can be used in the architecture. Moreover, the components can reside in a single computer environment or distributed across multiple computer environments.

FIG. 4A provides a more detailed depiction of the architecture of the ATB UI 102 according to one embodiment. As previously shown in FIG. 1, client computers 122 access the ATB UI 102 over one or more communication networks. The presentation component 110 presents a user interface by which users can interact with the ATB UI 102. Except for the web browsers located on client computers 122, no installed code base of ATB UI needs to reside on client computers 122. Communications from the browser interface to the presentation component 110 occur through URL navigation or HTTP requests.

The business logic abstraction layer 112 is an abstraction layer of the attribute/filter management business logic 114. The abstraction layer 110 further comprises a filter manager 132, an attribute manager 134, and a client manager 136. These managers present objects, and systems and methods for viewing, editing, and managing filters, attributes, and systems to which the filters and attributes will employ. The managers have their corresponding counterparts within the business logic 114, namely, a filter business component 142, an attribute business component 144, and a client business component 146. The three business components work with a repository object 120 within the data access layer 116 for transactions involving the repository 118. In this example embodiment, the repository 118 comprises a SQLite database server, but those skilled in the art will recognized that other database servers such as Oracle, mySQL, Microsoft SQL Server, and IBM DB2 can be used as well.

The abstraction layer 112 serves as a liaison between the presentation component 110 and the business logic 114. The abstraction layer 112 also provides security enforcement functions in one embodiment. For example, it ensures that attributes are only displayed for the users who have the proper access rights to view them. In one embodiment, the abstraction layer 112, the business logic 114 and the data access layer 116 are implemented in a Ruby on Rails environment, which has the advantages of providing ease of maintenance and a modular programming construct, though other environments may be used.

As shown in FIG. 4A, the ATB UI 102 further includes a core services component 254 and a support service component 256. In one embodiment, the core services component 254 is a Rails framework comprising a security component 260, an exception component 262, an export/import component 264, and a logger component 266. The security component 260 provides the logic for security enforcement, which limits access to the attributes and filters to those users who have the proper levels of access rights. The exception component 262 handles exceptions generated by any component within the ATB UI 102. The export/import component 264 is a data interchange service that supports and provides the capability to export and import ATB entities (for example, attributes) from and into the ATB UI 102. The logger component 266 is served by the Ruby log4r library in one embodiment. It provides the functionality to publish messages to external log stores and to control the type and volume of data logged. The support service component 256 comprises a reporting component 268, which provides a set of pre-defined reports that can be run. In one embodiment, the generated reports are templates for specification documents and include attributes and related results.

FIG. 4B is a schematic diagram showing the implementation of the ATB engine system 104 according to one embodiment. The façade 130 includes two methods of accepting requests. It has a web service component 270 and a batch component 272. In one embodiment, the web service component 270 is supported by a SOAP4R component 294 within technical services 288. SOAP4R is a Simple Object Access Protocol (SOAP) implementation in the Ruby programming language. The web service component 270 thus accepts SOAP requests from calling applications. Alternatively, the batch component 272 reads batched requests in a repository 298. In one embodiment, the repository 298 is a hard drive.

In either case, once an incoming request has been received, the façade 130 begins the process of managing the request and performs the necessary work in order to return a response to the request. First, the façade 130 has a message publisher 276 that will publish a message to either the ATB engine 106 or the data access engine 108. In one embodiment, the message is a RPC sent via the TCP/IP protocol. The façade 130 includes managing logic that will handle the sequence in which RPCs are sent. The façade 130 additionally includes a profile component that handles a variety of work-flow logic as described earlier.

Both the ATB engine 106 and the data access engine 108 have a message subscriber component that subscribes to the message publisher 276. In the ATB engine 106, once the message subscriber 278 receives a message from the message publisher 276, it will instruct business component 280 to perform calculations. In one embodiment, business component 280 further comprises the parser 162 and the business rules component 164. The business rules component 164 performs calculations based on the attributes and filters within deployment package 120, using data parsed by the parser 162. In one embodiment, the business component 280 is implemented in the Lua scripting language, which provides fast execution times and increased performance. This is particularly important in some embodiments because the ATB engine system may be configured to receive a high volume of requests from the batch component 272 and to accept a large number of requests from calling applications through the web service 270. The façade 130 and the access engine 108 are implemented in the Ruby programming language in one embodiment.

Access engine 108 also includes the aforementioned message subscriber 282 and a data source access framework 284. After the message subscriber 282 receives a message from the message publisher 276, it will instruct the data source access framework 284 to access the requested data. The data source access framework reads a data source definition file 286 in order to determine the data formats of the various data sources that are being accessed. Finally, technical services 288 include a log4r component that provides logging capability for debugging purposes.

Attribute Interfaces

FIG. 5A is an example computer screen showing the attribute search/browse interface 152 in the ATB UI 102 according to one embodiment. As shown, a search panel 302 is provided so users can search for attributes by various parameters such as name/keyword, system, and format. In response, the ATB UI 102 displays a sortable list 304 of attributes stored in the repository 118 that match the search parameters. For each attribute, the ATB UI 102 shows the system in which the attribute is being deployed, as well as whether the attribute is locked by another user who is viewing or editing it.

The ATB UI 102 facilitates the management of the attributes and removes some of the difficulties of having to manually configure the deployment of attributes. Because the UI provides a comprehensive view of how attributes are deployed in various systems, users can quickly and easily view the dependencies between attributes and the associated systems and determine how editing an attribute will affect the systems in their deployment scheme. In addition, because the name of each attribute listed is a link to the attribute development interface with that selected attribute shown in a coding area, the users can quickly launch into the development environment from the list 304.

FIG. 5B is an example computer screen showing how attributes can be edited in the attribute development interface 150 according to one embodiment. As shown, an example attribute is being edited or created in a coding area 314. Because attributes often reference numerous other attributes, filters, functions, and operators, the development interface 150 provides a panel 312 that lists other attributes, filters, functions, and operators that can be inserted into the coding area 314. This ensures that the newly created attribute or the edited attribute will have the proper dependencies. In one embodiment, the list is populated by the attributes and filters retrieved from the repository 118. Any attribute or filter listed can be moved (for example, dragged and dropped) into the coding area 314, and a proper reference to that attribute or filter automatically appears within the coding area 314.

In addition, the panel 312 has a tab area 316, which has two other tabs “filters” and “functions & operators” as shown. A user can select one of the other tabs and brings up either the filters or the functions and operators that are available. Likewise, these elements can be dragged and dropped into the coding area 314 and the proper references are automatically created. FIG. 5C shows an example of the panel 312 after a user selects the “functions & operators” tab within the tab area 316. Any of the functions and operators shown can be moved (for example, dragged and dropped) into the coding area 314. Therefore, the development interface 150 enables users to rapidly code new attributes or modify existing attributes with accurate references to other attributes, filters, functions, and operators.

Filter Interfaces

FIG. 6A is one embodiment of an example computer screen of the filter development interface 150. Similar to the attribute development interface shown in FIG. 5B, the filter development interface presents other filters, functions, and operators in a panel 322 that can be dragged and dropped into a coding area 324.

Coding Filters for Various Data Sources

Because of the complexities of database definitions, the task of developing filters on raw data is often left to programmers. To overcome this problem, the filter development interface 150 simplifies the task of developing filters for various data sources by showing the data fields within segments of the data sources. Therefore, in addition to filters, functions, and operators, the filter development interface has a “raw data” tab in the tab area 326. As shown in FIG. 6A, the available data segments in the data source “Experian” are listed in the panel 322. The fields within the segments are available to be dragged and dropped into the coding area 324. As shown, the coding area 324 shows the codes the filter that is being developed.

FIG. 6B is an example computer screen showing how filters can be developed for different data sources in the ATB UI. As shown in FIG. 6B, data fields for a selected segment within a selected data source are listed in the panel 322. Drop down boxes 320 include a data source drop-down box, a segment drop-down box, and an argument drop-down box. The data source is selected at the data source drop-down box, and the data segment is selected at the segment drop-down box and so on. In this example, the “Experian” data source has been selected and a segment has also been selected. In one embodiment, all available valid segments associated with the selected data source are listed in the segment drop-down box.

In one embodiment, once a segment is selected, all the available data fields are shown within the panel 322. The user can drag and drop a field into the coding area 324. Through the filter development interface 150, a user can easily scan through the available fields within the segments in a data source and quickly and accurately add the necessary fields to a filter that is currently being edited in the coding area 324. This interface minimizes errors and speeds up the development of filters for multiple data sources with different data formats.

FIG. 6C is a computer screen showing the filter search/browse interface 152. Much like the interface shown FIG. 5A for attributes, users can search for filters and the search results will be displayed in a listing 340. Each filter is listed with status information, whether it has been locked by another user for editing, and whether it has been encoded for a particular data source (as indicated by the check marks). Clicking on the name of the filter will launch a development interface similar to the one shown in FIG. 6A, where the user can edit the selected filter.

Syntax Checking and Validation

In one embodiment, ATB UI provides a syntax checking engine that checks the syntax of a filter or an attribute within a current coding area if requested by the user. As shown in FIG. 6A, once the user selects a “Check Syntax” button 330, the ATB UI will run a syntax check on either the filter or attribute that is currently being edited. For example, FIG. 6A shows the syntax check window 328 displaying a message indicating that the syntax check engine has discovered a syntax error on line 13 of the example filter. The syntax check engine can likewise check for syntax errors within an attribute. Thus users of the ATB UI can avoid saving invalid filters or attributes. In other embodiments, a syntax check is automatically performed when a user attempts to save a filter or an attribute to the repository 118, or save changes to an existing filter or attribute.

Besides a syntax checking engine, embodiments of ATB UI also provide the validation interface 156 and the validation engine 160 (shown in FIG. 1) through which users can validate attributes against test data. FIG. 7 shows an example computer screen of the validation interface 156. The validation interface 156 includes an information/selection area 350, which further includes a data file selector 352 and a data source selector 354. Using the data file selector 354, the user can select a sample test file including test data for the data source that the user selects in the data source selector 354. In the example shown, the user has selected a file including sample data from the data source “Experian” to be sent to the validation engine 160. In one embodiment, the validation engine 160 accepts test files for in the ARF format.

In the example shown in FIG. 7, the validation engine 160, through the validation interface 156, provides a result 358 for the selected attribute “ALL002.” The validation engine 160 has calculated the attribute “ALL002” against the test data and has determined that the result is 52.0. More importantly, the validation interface 156 further includes a validation table 356 in which a user can examine the raw data and the dependencies within the attribute that is undergoing validation. The column headings of the validation table 356 show all the depending filters and attributes referenced in the code of attribute “ALL002.” The rows of the validation table 356 show raw data from the test file.

The validation interface 156 advantageously allows the user to see exact results returned by each filter or attribute referenced by the attribute that is being validated. This enables the user to quickly determining any errors or performance problems hidden in the chain of dependencies. The user can, for example, quickly determine that a filter is selecting a wrong field or that an attribute calculation is returning erroneous or undesirable results. The validation interface 156 is particularly helpful in the development of aggregation attributes, which often reference a large number attributes and filters.

System Management

FIG. 8A is one embodiment of an example computer screen showing how a client can be created in the ATB UI. An input interface 360 solicits configuration information that is needed to create a new client. The input interface 360 includes a listing of data sources that the new client will access, as well as a listing of systems that will be associated with this new client. When the user attempts to save the client configuration information, the ATB UI will check for errors and display an error message 362 (as shown) if needed.

FIG. 8B is one embodiment of an example computer screen showing how attributes can be associated with a client system for deployment. As an example, a client who is a bank may wish to deploy various systems associated with its needs. The bank may thus have one system that screens potential home loan applicants and another system for finding prospective commercial credit card applicants. Because each system requires different data and calculations, the bank client may wish to configure each one differently. The ATB UI provides an interface 366 to allow a client user to associate various attributes to a system for deployment. The interface 366 includes an available attribute listing 368 and a deployment attribute listing 370. The user can search for and browse attributes that are available deployment in the listing 368 and then add the attributes to the deployment listing 370 by using an arrow selector 372. The attributes will then be slated for deployment for the particular system.

FIG. 8C is one embodiment of an example computer screen showing how multiple client system deployments are managed in the ATB UI. A system listing 380 lists systems that are available for deployment. Each system is listed with a name, a historical data entry including the date of last deployment, status, and the number of associated attributes. Under the status column, the ATB UI may indicate whether a system is ready or not ready for deployment. Some systems may not be ready to be deployed because they may have attributes with syntax errors or no attributes at all. The system listing 380 provides a way for the user to browse through the available systems and quickly determine which are ready for deployment. Furthermore, the ATB UI provides a link to the attributes so the user can quickly view or edit the attributes associated with the systems. The user is also given the option to select multiple systems from the listing 380 and deploy all of them at the same time. The deployment package 120 is then automatically created and can be used in ATB engine system 104.

Managing Data Sources and Attribute Exports

FIG. 9A is one embodiment of an example computer screen showing how multiple data sources are managed in the ATB UI 102. The ATB UI 102 provides a data source listing 386 that lists the available data sources. Each entry lists the short and long names of the data source as well as the time when the data source was last modified. The ATB UI 102 also provides a data source area 388 that displays the detail information of a data source that has been selected. The user can then view or edit the details of a particular data source, including the location of the file from which data can be obtained.

FIG. 9B is one embodiment of an example computer screen shot showing how attributes can be exported to a file. A system listing 390 provides a listing of systems that can be exported. Once exported, the file will include all attributes associated with the selected system. The export function allows multiple users from various sites to collaborate on a project. In one embodiment, users at remote sites without access to the centralized repository can run their own copies of the ATB UI 102. For example, a first user may be developing attributes and filters on a laptop while traveling. The export function enables the first user to export the attributes and filters and send the export file to a second user who runs a different instance of the ATB UI 102 that has a connection to a centralized repository. The second user will then be able to import the attributes created by the first user and save the attributes to the centralized repository. In one embodiment, the ATB UI 102 checks the import file against the repository, so that duplicate attributes are shown to the user. The second user will then have the option of overwriting the attributes that are in the repository with the imported version. In this manner, numerous users from various remote locations, even those without connection to a centralized attribute repository, can independently develop attributes and filters and merge their work later.

Various Embodiments of System and Method Implementations

In one embodiment, the systems and methods for generating and calculating attributes may be embodied in part or in whole in software that is running on one or more computing devices. The functionality provided for in the components and modules of the computing device(s) may comprise one or more components and/or modules. For example, the computing device(s) may comprise multiple central processing units (CPUs) and one or more mass storage device(s), such as may be implemented in an array of servers.

In general, the word “module,” “application”, or “engine,” as used herein, refers to logic embodied in hardware and/or firmware, and/or to a collection of software instructions, possibly having entry and exit points, written in a programming language, such as, for example, Java, Ruby, Ruby on Rails, Lua, C and/or C++. These may be compiled and linked into an executable program, installed in a dynamic link library, or may be written in an interpreted programming language such as, for example, BASIC, Perl, or Python. It will be appreciated that modules, applications, and engines may be callable from others and/or from themselves, and/or may be invoked in response to detected events or interrupts. Instructions may be embedded in firmware, such as an EPROM.

It will be further appreciated that hardware may be comprised of connected logic units, such as gates and flip-flops, and/or may be comprised of programmable units, such as programmable gate arrays or processors. The modules, applications, and engines described herein are in certain applications preferably implemented as software modules, but may be represented in hardware or firmware in other implementations. Generally, the modules, applications, and engines described herein refer to logical modules that may be combined with other modules and/or divided into sub-modules despite their physical organization or storage.

In some embodiments, the computing device(s) communicates with one or more databases that store information, including credit data and/or non-credit data. This database or databases may be implemented using a relational database, such as SQLite, Sybase, Oracle, CodeBase, mySQL, and Microsoft® SQL Server as well as other types of databases such as, for example, a flat file database, an entity-relationship database, and object-oriented database, and/or a record-based database.

In one embodiment, the computing device is IBM, Macintosh, and/or Linux/Unix compatible. In another embodiment, the computing device comprises a server, a laptop computer, a cell phone, a Blackberry, a personal digital assistant, a kiosk, or an audio player, for example. In one embodiment, the computing device includes one or more CPUs, which may each include microprocessors. The computing device may further include one or more memory devices, such as random access memory (RAM) for temporary storage of information and read only memory (ROM) for permanent storage of information, and one or more mass storage devices, such as hard drives, diskettes, or optical media storage devices. In one embodiment, the modules of the computing are in communication via a standards based bus system, such as bus systems using Peripheral Component Interconnect (PCI), Microchannel, SCSI, Industrial Standard Architecture (ISA) and Extended ISA (EISA) architectures, for example. In certain embodiments, components of the computing device communicate via a network, such as a local area network that may be secured.

The computing is generally controlled and coordinated by operating system software, such as the Windows 95, Windows 98, Windows NT, Windows 2000, Windows XP, Windows Vista, Linux, SunOS, Solaris, PalmOS, Blackberry OS, or other compatible operating systems. In Macintosh systems, the operating system may be any available operating system, such as MAC OS X. In other embodiments, the computing device may be controlled by a proprietary operating system. Conventional operating systems control and schedule computer processes for execution, perform memory management, provide file system, networking, and I/O services, and provide a user interface, such as a graphical user interface (GUI), among other things.

The computing device may include one or more commonly available input/output (I/O) devices and interfaces, such as a keyboard, mouse, touchpad, microphone, and printer. Thus, in one embodiment the computing device may be controlled using the keyboard and mouse input devices, while in another embodiment the user may provide voice commands to the computing device via a microphone. In one embodiment, the I/O devices and interfaces include one or more display device, such as a monitor, that allows the visual presentation of data to a user. More particularly, a display device provides for the presentation of GUIs, application software data, and multimedia presentations, for example. The computing device may also include one or more multimedia devices, such as speakers, video cards, graphics accelerators, and microphones, for example.

In one embodiment, the I/O devices and interfaces provide a communication interface to various external devices. For example, the computing device may be configured to communicate with one or more networks, such as any combination of one or more LANs, WANs, or the Internet, for example, via a wired, wireless, or combination of wired and wireless, communication links. The network communicates with various computing devices and/or other electronic devices via wired or wireless communication links.

Although the foregoing invention has been described in terms of certain embodiments, other embodiments will be apparent to those of ordinary skill in the art from the disclosure herein. Moreover, the described embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms without departing from the spirit thereof. Accordingly, other combinations, omissions, substitutions and modifications will be apparent to the skilled artisan in view of the disclosure herein. For purposes of discussing the invention, certain aspects, advantages and novel features of the invention have been described herein. Of course, it is to be understood that not necessarily all such aspects, advantages or features will be embodied in any particular embodiment of the invention.

Claims

1. A system for generating attributes, the system comprising: one or more processors; andnon-transitory computer storage comprising code executable by the one or more processors, the executable code causing the one or more processors to: receive, via a user interface, an indication of at least one attribute that relates to a type of credit data;identify a credit data attribute formula expression based on the indication of the at least one attribute;generate a deployment package that applies extracted data to the credit data attribute formula expression;receive at least one attribute score for the at least one attribute, the at least one attribute score generated by extracting first data associated with the type of credit data from a first data source, extracting second data associated with the type of credit data from a second data source, and using the deployment package to apply the extracted first and second data to the credit data attribute formula expression, the first data and the second data having different formats; anddisplay, via the user interface, the extracted first data, the extracted second data, the credit data attribute formula expression, or the at least one attribute score.

2. The system of claim 1, wherein the first data is extracted in accordance with a first filter and the second data is extracted in accordance with a second filter.

3. The system of claim 2, wherein the executable code further causes the one or more processors to display information in a format that allows identification of errors in one or more of the first filter, the second filter, and the credit data attribute formula expression.

4. The system of claim 1, wherein the at least one attribute is data-independent and wherein the first and second filters are data-dependent.

5. The system of claim 1, wherein the executable code further causes the one or more processors to display functions and operations associated with the first filter or functions and operations associated with the second filter.

6. The system of claim 1, wherein the executable code further causes the one or more processors to calculate a data score based at least in part on the at least one attribute score.

7. The system of claim 1, wherein the first filter is configured extract the first data by mapping one or more data fields in the first data to the type of credit data.

8. The system of claim 1, wherein the first data source comprises a first credit bureau data source and the second data source comprises a second credit bureau data source.

9. A computer-implemented method for generating attributes, the computer-implemented method comprising: receiving, via a user interface, an indication of at least one attribute that relates to a type of credit data;identifying a credit data attribute formula expression based on the indication of the at least one attribute;generating a deployment package that applies extracted data to the credit data attribute formula expression;receiving at least one attribute score for the at least one attribute, the at least one attribute score generated by extracting first data associated with the type of credit data from a first data source, extracting second data associated with the type of credit data from a second data source, and using the deployment package to apply the extracted first and second data to the credit data attribute formula expression, the first data and the second data having different formats; anddisplay, via the user interface, the extracted first data, the extracted second data, the credit data attribute formula expression, or the at least one attribute score.

10. The computer-implemented method of claim 9, wherein the first data is extracted in accordance with a first filter and the second data is extracted in accordance with a second filter.

11. The computer-implemented method of claim 10 further comprising displaying information in a format that allows identification of errors in one or more of the first filter, the second filter, and the credit data attribute formula expression.

12. The computer-implemented method of claim 9, wherein the at least one attribute is data-independent and wherein the first and second filters are data-dependent.

13. The computer-implemented method of claim 9 further comprising displaying functions and operations associated with the first filter or functions and operations associated with the second filter.

14. The computer-implemented method of claim 9 further comprising calculating a data score based at least in part on the at least one attribute score.

15. The computer-implemented method of claim 9, wherein the first filter is configured extract the first data by mapping one or more data fields in the first data to the type of credit data.

16. A non-transitory computer storage medium which stores a client application comprising executable code, the executable code causing a computing device to: receive, via a user interface, an indication of at least one attribute that relates to a type of credit data;identify a credit data attribute formula expression based on the indication of the at least one attribute;generate a deployment package that applies extracted data to the credit data attribute formula expression;receive at least one attribute score for the at least one attribute, the at least one attribute score generated by extracting first data associated with the type of credit data from a first data source, extracting second data associated with the type of credit data from a second data source, and using the deployment package to apply the extracted first and second data to the credit data attribute formula expression, the first data and the second data having different formats; anddisplay, via the user interface, the extracted first data, the extracted second data, the credit data attribute formula expression, or the at least one attribute score.

17. The non-transitory computer storage medium of claim 16, wherein the first data source is selected from the group consisting of a credit bureau source, a non-credit bureau source, and a local data source.

18. The non-transitory computer storage medium of claim 16, wherein the first data is extracted in accordance with a first filter and the second data is extracted in accordance with a second filter.

19. The non-transitory computer storage medium of claim 16, wherein the executable code further causes the computing device to display information in a format that allows identification of errors in one or more of the first filter, the second filter, and the credit data attribute formula expression.

20. The non-transitory computer storage medium of claim 16, wherein the executable code further causes the computing device to calculate a data score based at least in part on the at least one attribute score.