Decision Tree Model Training for Automated Loan Adjudication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The current loan adjudication process is inefficient, requiring significant manual work and time, typically taking days or weeks to complete.
Innovation Solution
A system that uses a machine learning model, specifically a decision tree, to automatically adjudicate loan applications by extracting user attributes, querying a machine learning model via an API, and determining approval based on identified rules, thereby reducing the need for human intervention and accelerating the decision-making process to a few seconds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual review process is used, then accuracy of loan decision can be maintained, but time required for adjudication increases significantly
Solution Approach 1:
The patent replaces the mechanical manual review system with an automated machine learning decision tree system. The decision tree model, trained on historical loan data, automatically evaluates applicant attributes and makes adjudication decisions without human intervention, thereby reducing time from days/weeks to seconds while maintaining decision accuracy through systematic rule-based evaluation.
Solution Approach 2:
The patent transforms the loan adjudication process by changing the evaluation parameters from subjective human judgment to objective data-driven criteria. The decision tree uses structured attributes (income, credit score, debt-to-income ratio, etc.) and applies predefined thresholds and rules to automatically determine loan approval, eliminating the time-consuming nature of manual review while preserving decision reliability.
2Productivity
If automated machine learning model is used, then processing speed increases, but complexity of the system increases
Solution Approach 1:
The patent segments the complex loan adjudication process into distinct, manageable components represented as nodes in a decision tree. Each node evaluates a specific attribute (e.g., credit score, income level, debt-to-income ratio) against predefined thresholds, breaking down the overall complex decision-making process into simpler, independent evaluation steps that can be processed automatically and efficiently.
3Ease of operation
If manual adjudication process is used, then transparency of decision-making can be maintained through human oversight, but productivity decreases
Solution Approach 1:
The patent incorporates feedback mechanisms that provide transparent explanations for automated loan decisions. The system generates detailed reports showing which attributes were evaluated, what thresholds were applied, and how the final decision was reached, allowing stakeholders to understand and audit the decision-making process even though it occurs automatically without human intervention.
Data Source
AI summary
An example operation may include one or more of assigning different criteria to a plurality of nodes of a decision tree model, respectively, iteratively executing the decision tree model on a plurality of training data which causes the plurality of training data to be assigned to the plurality of nodes of the decision tree model based on the different criteria assigned to the plurality of nodes, identifying nodes among the plurality of nodes within the decision tree model which have a purity above a predetermined purity threshold, generating a set of rules based on the identified nodes which have the purity above the predetermined purity threshold, and embedding the set of rules within the decision tree model and storing the decision tree model within a storage device.