Nice has been granted a patent for a computerized method that enhances financial crime detection accuracy in Machine Learning models. The method involves retrieving and processing financial transaction records, performing feature engineering, and generating complex datasets to improve classification and risk scoring from the model’s initiation stage. GlobalData’s report on Nice gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Nice, Intelligent contact centers was a key innovation area identified from patents. Nice's grant share as of June 2024 was 60%. Grant share is based on the ratio of number of grants to total number of patents.
Method for improving financial crime detection using machine learning
The patent US12045840B2 outlines a computerized method and system designed to enhance the accuracy of financial crime detection through machine learning (ML) from the initial stages of model implementation. Central to this innovation is a Representative Dataset Generation (RDG) module that retrieves and processes financial transaction records from a database. The method involves feature engineering to create probabilistic categorical and numerical features, which are then combined to form a complex features dataset. This dataset is utilized by an ML model to classify financial transactions based on risk scores. Transactions with scores below a specified threshold are processed further, while those above the threshold trigger alerts for manual evaluation by analysts. The RDG module also adapts the ML model through supervised model tuning, training, and validation, thereby improving classification accuracy from the outset.
Additionally, the patent specifies that the RDG module performs data validation and fraud tagging to ensure the completeness and reliability of the financial transaction records. The model tuning, training, and testing processes can also be executed using an Anomaly Detection Model (ADM) based on the Isolation Forest algorithm. Various sampling techniques, including random, time-based, and stratified sampling, are employed to enhance the dataset's representativeness. This comprehensive approach aims to refine the detection of financial crimes, ensuring that the system can effectively identify and respond to potential fraudulent activities in real-time.
To know more about GlobalData’s detailed insights on Nice, 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.