Channel Affinity Mapping for Frictionless Transaction Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing enterprise systems require manual login and navigation through multiple steps on a new channel to complete a transaction not supported on the incoming channel, leading to inefficiency and increased time consumption.
Innovation Solution
An enterprise computing platform uses an eligibility machine learning model to determine a preferred channel based on customer unique identifiers and transaction intent, generating a deep link that automatically authenticates and completes the transaction on the preferred channel without manual login.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual login and navigation through multiple steps is required on a new channel, then the system can complete transactions on channels not supported by the incoming channel, but the transaction time increases and efficiency decreases
Solution Approach 1:
The system performs preliminary actions by pre-extracting features from channel input data, pre-processing channel information, and pre-determining preferred channels before the actual transaction occurs. The eligibility machine learning model is pre-trained on historical channel input data to quickly determine the preferred channel without requiring manual navigation steps during transaction execution.
Solution Approach 2:
The patent introduces an intermediary eligibility machine learning model that acts as a mediator between the incoming channel and the preferred channel. This model processes numerical features and intent features from channel input data to determine the preferred channel, enabling automatic channel switching without manual user intervention and reducing transaction time.
2Reliability
If manual login and navigation steps are required on the new channel, then the system can ensure proper authentication, but the ease of operation decreases
Solution Approach 1:
The system implements self-service by having the eligibility machine learning model automatically determine the preferred channel and generate deep links without requiring manual user input. The deep link automatically handles authentication and navigation, allowing the system to serve itself in determining the optimal channel while maintaining security through automated authentication mechanisms.
Solution Approach 2:
The patent replaces the mechanical manual login and navigation process with an automated electronic system. The eligibility machine learning model substitutes manual channel selection with automated feature extraction and processing, while deep links automatically handle authentication, eliminating the need for users to manually navigate through multiple steps while maintaining secure authentication.
3Measurement precision
If the system extracts and processes features from channel input data using machine learning, then the accuracy of preferred channel determination improves, but the device complexity increases
Solution Approach 1:
The system segments the complex channel selection process into distinct manageable components: feature extraction modules that extract numerical features from channel input data, intent feature extraction modules that process intent tags, and the eligibility machine learning model that combines these features. This segmentation allows each component to be optimized independently while working together to achieve accurate channel determination.
Solution Approach 2:
The patent applies parameter changes by transforming channel input data into different feature representations. The system extracts numerical features from channel input data and intent features from intent tags, then combines these transformed parameters in the eligibility machine learning model. This parameter transformation enables accurate channel selection by converting raw data into meaningful features that the model can process effectively.
Data Source
AI summary
Systems and methods are disclosed for real time, frictionless channel switching to map and transpose a transaction on a preferred channel from an incoming channel. The systems and methods may use a machine learning model that processes features from a plurality of channels to determine the preferred channel. An omni channel processor is used to determine intended functionality data of the transaction for the preferred channel. The systems and methods may generate transposed intent data based on the intended functionality data of the transaction for the preferred channel. A deep link is generated and embedded with the transposed intent data and a channel access token for the transaction on the preferred channel. The deep link is configured to access and complete the requested transaction on the preferred channel.