Customer Data Prioritization Using Prompted Vector Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning models struggle to effectively process and prioritize tabular customer datasets, which are commonly used in contact centers and financial institutions, leading to suboptimal performance in identifying and addressing issues such as fraudulent transactions.
Innovation Solution
A method and system that utilize machine learning models to generate vectors from tabular data, applying analysis prompts to prioritize customer datasets by comparing prioritization values to threshold values, and generating alerts for high-risk transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine learning models are used to process tabular customer data, then the ability to identify high-risk transactions is improved, but the complexity of the system increases
Solution Approach 1:
The patent introduces an intermediary conversion layer that transforms tabular data into text format, which then serves as input to machine learning models. This mediator (text conversion layer) simplifies the interaction between tabular data and ML models, reducing system complexity while maintaining identification accuracy.
Solution Approach 2:
The system converts tabular data into text format that can be processed by universal language models. This multi-functional approach allows the same text-processing infrastructure to handle various tabular data types, reducing the need for separate specialized processing systems for different data formats.
2Ease of operation
If tabular data is converted to text format for ML processing, then the ease of operation is improved, but the loss of information may increase
Solution Approach 1:
The conversion process segments the tabular data into meaningful text representations that preserve key information. By dividing the data into structured text elements that maintain relationships between rows and columns, the system preserves information while achieving ease of text-based processing.
Solution Approach 2:
The system performs preliminary text conversion of tabular data before feeding it to machine learning models. This preliminary action prepares the data in a format that is easier to process while preserving essential information structures, allowing subsequent ML processing to operate more effectively.
3Productivity
If machine learning models process large amounts of tabular data, then the productivity is improved, but the use of energy increases
Solution Approach 1:
The system extracts and converts only the essential information from large tabular datasets into text format before processing. By extracting only necessary data elements and converting them to text, the system reduces the volume of data requiring ML processing, thereby improving productivity while reducing energy consumption.
Solution Approach 2:
The system applies partial conversion of tabular data to text format, converting only the portions that are most useful for ML processing. This partial action approach avoids the excessive energy consumption of converting entire datasets while still providing sufficient information for effective machine learning processing.
Data Source
AI summary
A system and method for prioritizing customer data may include a computing device; a memory; and a processor, the processor configured to: use of one or more datasets of tabular customer data to generate one or more analysis prompts; apply the one or more analysis prompts to a machine learning model to generate a vector; and generate a prioritization of the one or more customer datasets by comparing a prioritization value of the vector to threshold values.