Discover Financial Services has been granted a patent for a machine learning model interpretability framework. This method groups input variables based on correlations, generates partial dependence plots for each group, and calculates contributions to model outputs, enhancing the understanding of variable impacts on classification decisions. GlobalData’s report on Discover Financial Services gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
According to GlobalData’s company profile on Discover Financial Services, UWB-enabled payments was a key innovation area identified from patents. Discover Financial Services's grant share as of July 2024 was 36%. Grant share is based on the ratio of number of grants to total number of patents.
Machine learning model interpretability using variable grouping and scoring
The patent US12050975B2 outlines a method and system for enhancing machine learning model interpretability through the use of partial dependence plots (PDPs). The method involves training a machine learning model on a dataset of input vectors and corresponding output scores. It begins by evaluating correlations among input variables to categorize them into groups. For each group, a PDP table is constructed, which includes a grid of sample points and their associated PDP values, indicating the contribution of each variable group to the model's output. Once the model is trained and the PDP tables are established, the system can receive new input vectors, generate scores, and make classification decisions based on these scores.
Additionally, the method allows for the determination of each variable group's contribution to the score without directly using the PDP function. This is achieved by accessing the respective PDP table, identifying neighboring sample points, and performing interpolation to derive an interpolated PDP value. The system can also generate adverse action reason codes when the classification decision involves credit extension, providing insights into which variable groups influenced the decision. The patent further details the use of various interpolation techniques and the potential integration of Shapley Additive Explanation (SHAP) analysis to enhance the understanding of variable contributions, thereby improving transparency in machine learning applications, particularly in financial decision-making contexts.
To know more about GlobalData’s detailed insights on Discover Financial Services, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
