Decision Engine
Decision Engine
Decision Engine
Decision Engine Core: Elements and Concepts
Decision Engine Core: Elements and Concepts
Jan 26, 2024
Jan 26, 2024
Jan 26, 2024
A decision engine should have a few fundamental core concepts. Such as the ability to construct rules with multiple outcomes, integrated with third party data providers and an open API to take in datasets from a variety of different sources. Some decision engines are messy to work with and require many engineers and analysts to implement rules and ingest data. Some other decision engines have a great deal of authoring capabilities but there’s no integration point to connect to other enterprise software.
Decision Engine Core: Elements and Concepts
A decision engine should have a few fundamental core concepts. Such as the ability to construct rules with multiple outcomes, integrated with third party data providers and an open API to take in datasets from a variety of different sources. Some decision engines are messy to work with and require many engineers and analysts to implement rules and ingest data. Some other decision engines have a great deal of authoring capabilities but there’s no integration point to connect to other enterprise software.
LendAPI’s decision engine is a standalone application as well as an fully integrated system that connects to third party data providers, customer communication channels such as email, sms and in-portal alerts. Rule sets are also connected directly to customer journey application flow and executes rules in real time.
Today we are going to take a look at some Decision Engine Core concepts and how LendAPI’s visual decision engine core can accomplish everything that you need to launch your banking and financing product.
Decision Engine Core: Variables, Actions and Outcomes
LendAPI's rules engine introduces three key concepts: variables, actions and outcomes.
Variables: These are the building blocks of rules, representing the data elements on which rules are applied. Users can define variables based on their specific use case, providing the flexibility to adapt the rules engine to various scenarios. Variables can be defined from a user input form, or the variables from a credit bureau API, or any other data providers via API. Here are the variable types:
Numeric, it can be an integer, float
String, a combination of letters, numbers or special characters
Boolean, it can be True or False
Date, this is the date in mm/dd/yyyy format
List, list data type is for a list of predefined options by the user, from which only one option can be selected at a time.
LendAPI also allows designs of custom variables and mathematical operations where clients can design their own variables to be used in underwriting rules or display these special variables as part of the application journey. This powerful tool gives our clients the ability to design interest rulesets that’s unique to their own design.
Actions: There are four types of actions: approve, decline, review and request_doc. Approve indicates the application is approved. Decline means the application is declined. Review indicates that the application is in manual review. Request_doc means the application needs further document uploads.
These actions can be used at the end of a decision node to indicate how you want your workflow to act or react.
Outcomes: On each action, the decision engine can be further drilled down to “outcome”. The outcome of a rule is the result produced when the conditions are met. LendAPI's rules engine allows users to clearly define and understand the potential outcomes, enhancing transparency and decision-making. For example, for an approved application, you may generate super prime, prime, sub-prime as the product outcomes.
Decision Engine Core: Visual Rules Editor in LendAPI
LendAPI's rules engine distinguishes itself by incorporating an easy to use and drag and drop visual rules editor. This user-friendly interface allows both technical and non-technical users to create, modify, and manage rules without delving into complex code. The visual editor simplifies the rule definition process, making it more accessible and efficient.
In the following screenshot, we define a rule which incorporates the variable “age”. When age is less than 18 years old, the application outcome is set as declined. When “age” is greater or equal than 18, the decision engine goes to the next node.
Decision Engine Core: Rule testing procedures
LendAPI provides a testing tool for the decision engine. Simply click the "Test" button within the visual engine editor to validate and test your rules.
In the example below, the outcome for an age of 17 is depicted. The action is marked as declined, accompanied by a specific outcome, often associated with an adverse action notification.
Decision Engine Core: Rules Versioning
Effective version control is a critical aspect of managing rules engines, ensuring traceability, collaboration, and the ability to rollback to previous states. Version control allows organizations to keep track of changes made to rules, understand the evolution of decision logic, and maintain a reliable history for auditing purposes.
In the context of rules engines, version control involves capturing snapshots of rule sets at different points in time. Each snapshot represents a version, and changes made to rules are associated with specific versions.
Once the decision tree is constructed using the visual rules editor, you have the option to publish the rules to LendAPI's integrated rules versioning system. Upon publication, the decision tree is preserved as a read-only copy, creating a distinct version. This versioned decision tree becomes a valuable asset for application decision processes, offering a stable and reproducible set of rules. Furthermore, the versioning system facilitates the implementation of A/B tests based on different decision tree versions. This empowers users to experiment with variations, analyze the impact on outcomes, and make informed decisions to optimize rule-based processes.
Decision Engine Core: Rules Engine API
LendAPI's decision tree functionality is seamlessly integrated with a RESTful API, offering a robust set of endpoints that empower users to interact with decision trees programmatically. This API provides a suite of operations, enabling users to perform various tasks such as listing decision trees, displaying decision tree versions, listing decision tree parameters, and executing decision trees. Check out the API documentation at https://lendapi.readme.io/reference/list-all-decision-trees
Listing Decision Trees:
The RESTful API allows users to retrieve a comprehensive list of decision trees within the LendAPI environment. This operation is particularly valuable for gaining an overview of existing decision trees, facilitating efficient management and navigation through the available rule structures.
Displaying Decision Tree Versions:
Users can leverage the API to access and display different versions of a specific decision tree. This capability ensures visibility into the evolution of decision logic over time. Versioned decision trees are invaluable for maintaining a historical record, auditing changes, and ensuring traceability in decision-making processes.
Listing Decision Tree Parameters:
The API provides the ability to retrieve a list of parameters associated with a decision tree. Decision tree parameters are key inputs that influence the outcome of the decision-making process. Listing these parameters programmatically enables users to understand and manage the inputs that guide the decision tree's behavior.
Running a Decision Tree:
Perhaps one of the most crucial features, the API allows users to execute a decision tree, triggering the evaluation of rules and obtaining the resulting decision. This dynamic functionality is vital for integrating decision-making processes into applications, workflows, or other systems where real-time decision outcomes are needed.
By exposing these decision tree operations through a RESTful API, LendAPI empowers developers, data scientists, and decision-makers to seamlessly integrate and automate decision logic within their applications. This level of programmability not only enhances the flexibility of rule-based systems but also facilitates the incorporation of decision trees into diverse and dynamic business processes.
Try It Yourself: Register on LendAPI.com
Curious to experience the power of a rules engine with a visual editor? Head over to LendAPI.com and try registering to witness firsthand how intuitive rule creation can be.
In conclusion, a rules engine with a visual editor democratizes the rule creation process, enabling a wider range of users to actively participate in defining and modifying business rules. LendAPI's approach simplifies the complexity associated with traditional rules engines, making it a valuable tool for businesses seeking agility and efficiency in their decision-making processes.
A decision engine should have a few fundamental core concepts. Such as the ability to construct rules with multiple outcomes, integrated with third party data providers and an open API to take in datasets from a variety of different sources. Some decision engines are messy to work with and require many engineers and analysts to implement rules and ingest data. Some other decision engines have a great deal of authoring capabilities but there’s no integration point to connect to other enterprise software.
Decision Engine Core: Elements and Concepts
A decision engine should have a few fundamental core concepts. Such as the ability to construct rules with multiple outcomes, integrated with third party data providers and an open API to take in datasets from a variety of different sources. Some decision engines are messy to work with and require many engineers and analysts to implement rules and ingest data. Some other decision engines have a great deal of authoring capabilities but there’s no integration point to connect to other enterprise software.
LendAPI’s decision engine is a standalone application as well as an fully integrated system that connects to third party data providers, customer communication channels such as email, sms and in-portal alerts. Rule sets are also connected directly to customer journey application flow and executes rules in real time.
Today we are going to take a look at some Decision Engine Core concepts and how LendAPI’s visual decision engine core can accomplish everything that you need to launch your banking and financing product.
Decision Engine Core: Variables, Actions and Outcomes
LendAPI's rules engine introduces three key concepts: variables, actions and outcomes.
Variables: These are the building blocks of rules, representing the data elements on which rules are applied. Users can define variables based on their specific use case, providing the flexibility to adapt the rules engine to various scenarios. Variables can be defined from a user input form, or the variables from a credit bureau API, or any other data providers via API. Here are the variable types:
Numeric, it can be an integer, float
String, a combination of letters, numbers or special characters
Boolean, it can be True or False
Date, this is the date in mm/dd/yyyy format
List, list data type is for a list of predefined options by the user, from which only one option can be selected at a time.
LendAPI also allows designs of custom variables and mathematical operations where clients can design their own variables to be used in underwriting rules or display these special variables as part of the application journey. This powerful tool gives our clients the ability to design interest rulesets that’s unique to their own design.
Actions: There are four types of actions: approve, decline, review and request_doc. Approve indicates the application is approved. Decline means the application is declined. Review indicates that the application is in manual review. Request_doc means the application needs further document uploads.
These actions can be used at the end of a decision node to indicate how you want your workflow to act or react.
Outcomes: On each action, the decision engine can be further drilled down to “outcome”. The outcome of a rule is the result produced when the conditions are met. LendAPI's rules engine allows users to clearly define and understand the potential outcomes, enhancing transparency and decision-making. For example, for an approved application, you may generate super prime, prime, sub-prime as the product outcomes.
Decision Engine Core: Visual Rules Editor in LendAPI
LendAPI's rules engine distinguishes itself by incorporating an easy to use and drag and drop visual rules editor. This user-friendly interface allows both technical and non-technical users to create, modify, and manage rules without delving into complex code. The visual editor simplifies the rule definition process, making it more accessible and efficient.
In the following screenshot, we define a rule which incorporates the variable “age”. When age is less than 18 years old, the application outcome is set as declined. When “age” is greater or equal than 18, the decision engine goes to the next node.
Decision Engine Core: Rule testing procedures
LendAPI provides a testing tool for the decision engine. Simply click the "Test" button within the visual engine editor to validate and test your rules.
In the example below, the outcome for an age of 17 is depicted. The action is marked as declined, accompanied by a specific outcome, often associated with an adverse action notification.
Decision Engine Core: Rules Versioning
Effective version control is a critical aspect of managing rules engines, ensuring traceability, collaboration, and the ability to rollback to previous states. Version control allows organizations to keep track of changes made to rules, understand the evolution of decision logic, and maintain a reliable history for auditing purposes.
In the context of rules engines, version control involves capturing snapshots of rule sets at different points in time. Each snapshot represents a version, and changes made to rules are associated with specific versions.
Once the decision tree is constructed using the visual rules editor, you have the option to publish the rules to LendAPI's integrated rules versioning system. Upon publication, the decision tree is preserved as a read-only copy, creating a distinct version. This versioned decision tree becomes a valuable asset for application decision processes, offering a stable and reproducible set of rules. Furthermore, the versioning system facilitates the implementation of A/B tests based on different decision tree versions. This empowers users to experiment with variations, analyze the impact on outcomes, and make informed decisions to optimize rule-based processes.
Decision Engine Core: Rules Engine API
LendAPI's decision tree functionality is seamlessly integrated with a RESTful API, offering a robust set of endpoints that empower users to interact with decision trees programmatically. This API provides a suite of operations, enabling users to perform various tasks such as listing decision trees, displaying decision tree versions, listing decision tree parameters, and executing decision trees. Check out the API documentation at https://lendapi.readme.io/reference/list-all-decision-trees
Listing Decision Trees:
The RESTful API allows users to retrieve a comprehensive list of decision trees within the LendAPI environment. This operation is particularly valuable for gaining an overview of existing decision trees, facilitating efficient management and navigation through the available rule structures.
Displaying Decision Tree Versions:
Users can leverage the API to access and display different versions of a specific decision tree. This capability ensures visibility into the evolution of decision logic over time. Versioned decision trees are invaluable for maintaining a historical record, auditing changes, and ensuring traceability in decision-making processes.
Listing Decision Tree Parameters:
The API provides the ability to retrieve a list of parameters associated with a decision tree. Decision tree parameters are key inputs that influence the outcome of the decision-making process. Listing these parameters programmatically enables users to understand and manage the inputs that guide the decision tree's behavior.
Running a Decision Tree:
Perhaps one of the most crucial features, the API allows users to execute a decision tree, triggering the evaluation of rules and obtaining the resulting decision. This dynamic functionality is vital for integrating decision-making processes into applications, workflows, or other systems where real-time decision outcomes are needed.
By exposing these decision tree operations through a RESTful API, LendAPI empowers developers, data scientists, and decision-makers to seamlessly integrate and automate decision logic within their applications. This level of programmability not only enhances the flexibility of rule-based systems but also facilitates the incorporation of decision trees into diverse and dynamic business processes.
Try It Yourself: Register on LendAPI.com
Curious to experience the power of a rules engine with a visual editor? Head over to LendAPI.com and try registering to witness firsthand how intuitive rule creation can be.
In conclusion, a rules engine with a visual editor democratizes the rule creation process, enabling a wider range of users to actively participate in defining and modifying business rules. LendAPI's approach simplifies the complexity associated with traditional rules engines, making it a valuable tool for businesses seeking agility and efficiency in their decision-making processes.
A decision engine should have a few fundamental core concepts. Such as the ability to construct rules with multiple outcomes, integrated with third party data providers and an open API to take in datasets from a variety of different sources. Some decision engines are messy to work with and require many engineers and analysts to implement rules and ingest data. Some other decision engines have a great deal of authoring capabilities but there’s no integration point to connect to other enterprise software.
Decision Engine Core: Elements and Concepts
A decision engine should have a few fundamental core concepts. Such as the ability to construct rules with multiple outcomes, integrated with third party data providers and an open API to take in datasets from a variety of different sources. Some decision engines are messy to work with and require many engineers and analysts to implement rules and ingest data. Some other decision engines have a great deal of authoring capabilities but there’s no integration point to connect to other enterprise software.
LendAPI’s decision engine is a standalone application as well as an fully integrated system that connects to third party data providers, customer communication channels such as email, sms and in-portal alerts. Rule sets are also connected directly to customer journey application flow and executes rules in real time.
Today we are going to take a look at some Decision Engine Core concepts and how LendAPI’s visual decision engine core can accomplish everything that you need to launch your banking and financing product.
Decision Engine Core: Variables, Actions and Outcomes
LendAPI's rules engine introduces three key concepts: variables, actions and outcomes.
Variables: These are the building blocks of rules, representing the data elements on which rules are applied. Users can define variables based on their specific use case, providing the flexibility to adapt the rules engine to various scenarios. Variables can be defined from a user input form, or the variables from a credit bureau API, or any other data providers via API. Here are the variable types:
Numeric, it can be an integer, float
String, a combination of letters, numbers or special characters
Boolean, it can be True or False
Date, this is the date in mm/dd/yyyy format
List, list data type is for a list of predefined options by the user, from which only one option can be selected at a time.
LendAPI also allows designs of custom variables and mathematical operations where clients can design their own variables to be used in underwriting rules or display these special variables as part of the application journey. This powerful tool gives our clients the ability to design interest rulesets that’s unique to their own design.
Actions: There are four types of actions: approve, decline, review and request_doc. Approve indicates the application is approved. Decline means the application is declined. Review indicates that the application is in manual review. Request_doc means the application needs further document uploads.
These actions can be used at the end of a decision node to indicate how you want your workflow to act or react.
Outcomes: On each action, the decision engine can be further drilled down to “outcome”. The outcome of a rule is the result produced when the conditions are met. LendAPI's rules engine allows users to clearly define and understand the potential outcomes, enhancing transparency and decision-making. For example, for an approved application, you may generate super prime, prime, sub-prime as the product outcomes.
Decision Engine Core: Visual Rules Editor in LendAPI
LendAPI's rules engine distinguishes itself by incorporating an easy to use and drag and drop visual rules editor. This user-friendly interface allows both technical and non-technical users to create, modify, and manage rules without delving into complex code. The visual editor simplifies the rule definition process, making it more accessible and efficient.
In the following screenshot, we define a rule which incorporates the variable “age”. When age is less than 18 years old, the application outcome is set as declined. When “age” is greater or equal than 18, the decision engine goes to the next node.
Decision Engine Core: Rule testing procedures
LendAPI provides a testing tool for the decision engine. Simply click the "Test" button within the visual engine editor to validate and test your rules.
In the example below, the outcome for an age of 17 is depicted. The action is marked as declined, accompanied by a specific outcome, often associated with an adverse action notification.
Decision Engine Core: Rules Versioning
Effective version control is a critical aspect of managing rules engines, ensuring traceability, collaboration, and the ability to rollback to previous states. Version control allows organizations to keep track of changes made to rules, understand the evolution of decision logic, and maintain a reliable history for auditing purposes.
In the context of rules engines, version control involves capturing snapshots of rule sets at different points in time. Each snapshot represents a version, and changes made to rules are associated with specific versions.
Once the decision tree is constructed using the visual rules editor, you have the option to publish the rules to LendAPI's integrated rules versioning system. Upon publication, the decision tree is preserved as a read-only copy, creating a distinct version. This versioned decision tree becomes a valuable asset for application decision processes, offering a stable and reproducible set of rules. Furthermore, the versioning system facilitates the implementation of A/B tests based on different decision tree versions. This empowers users to experiment with variations, analyze the impact on outcomes, and make informed decisions to optimize rule-based processes.
Decision Engine Core: Rules Engine API
LendAPI's decision tree functionality is seamlessly integrated with a RESTful API, offering a robust set of endpoints that empower users to interact with decision trees programmatically. This API provides a suite of operations, enabling users to perform various tasks such as listing decision trees, displaying decision tree versions, listing decision tree parameters, and executing decision trees. Check out the API documentation at https://lendapi.readme.io/reference/list-all-decision-trees
Listing Decision Trees:
The RESTful API allows users to retrieve a comprehensive list of decision trees within the LendAPI environment. This operation is particularly valuable for gaining an overview of existing decision trees, facilitating efficient management and navigation through the available rule structures.
Displaying Decision Tree Versions:
Users can leverage the API to access and display different versions of a specific decision tree. This capability ensures visibility into the evolution of decision logic over time. Versioned decision trees are invaluable for maintaining a historical record, auditing changes, and ensuring traceability in decision-making processes.
Listing Decision Tree Parameters:
The API provides the ability to retrieve a list of parameters associated with a decision tree. Decision tree parameters are key inputs that influence the outcome of the decision-making process. Listing these parameters programmatically enables users to understand and manage the inputs that guide the decision tree's behavior.
Running a Decision Tree:
Perhaps one of the most crucial features, the API allows users to execute a decision tree, triggering the evaluation of rules and obtaining the resulting decision. This dynamic functionality is vital for integrating decision-making processes into applications, workflows, or other systems where real-time decision outcomes are needed.
By exposing these decision tree operations through a RESTful API, LendAPI empowers developers, data scientists, and decision-makers to seamlessly integrate and automate decision logic within their applications. This level of programmability not only enhances the flexibility of rule-based systems but also facilitates the incorporation of decision trees into diverse and dynamic business processes.
Try It Yourself: Register on LendAPI.com
Curious to experience the power of a rules engine with a visual editor? Head over to LendAPI.com and try registering to witness firsthand how intuitive rule creation can be.
In conclusion, a rules engine with a visual editor democratizes the rule creation process, enabling a wider range of users to actively participate in defining and modifying business rules. LendAPI's approach simplifies the complexity associated with traditional rules engines, making it a valuable tool for businesses seeking agility and efficiency in their decision-making processes.