Structured Data Tokenization for Large Language Model Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large language models struggle with input data that is not in natural language format, leading to poor quality output and unobservable internal workings, making them difficult to use effectively.
Innovation Solution
Process event and tabular data into sequential data tokens, embeddings, and other forms suitable for input to large language models using techniques like alphanumeric transformation, template, embedding projection, clustering, and decision trees to improve data suitability and output quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If event data and tabular data are provided directly to large language models without processing, then the input process is simple, but the output quality becomes poor and the model cannot effectively process non-natural language data
Solution Approach 1:
The patent applies preliminary action by transforming event data and tabular data into sequential token formats before they are input to the large language model. This preprocessing step converts structured data into the natural language sequence format that the model expects, thereby improving output quality without requiring changes to the model itself. The transformation includes mapping data fields to tokens and organizing them in sequential order that the model can process effectively.
Solution Approach 2:
The patent introduces an intermediary data transformation layer between the structured data source and the large language model. This intermediary component acts as a mediator that converts event data and tabular data into sequential token representations, enabling the model to process non-natural language data effectively. The intermediary transformation process includes field mapping, tokenization, and sequence construction, which bridges the gap between structured data and the model's natural language processing capabilities.
2Adaptability or versatility
If large language models are used for processing structured data, then linguistic tasks can be performed, but the inner workings become unobservable and difficult to use effectively
Solution Approach 1:
The patent applies feedback by implementing a classification task that provides observable output from the large language model processing. The system takes processed sequential data, passes it through the model, and generates classification results that can be observed and measured. This feedback mechanism makes the model's decision-making process more transparent and usable for practical applications such as fraud detection, while maintaining the model's versatility in performing linguistic and classification tasks.
Data Source
AI summary
Aspects described herein may relate to techniques and/or methods that process certain forms of event data and/or tabular data for input to one or more machine learning models, such as a large language model. Additional aspects may relate to using the output of a large language model as part of a process for detecting fraud based on the event data and/or tabular data. In some variations, the event data and/or tabular data may be processed into data tokens, embeddings, or other forms of data suitable for use as input to a large language model.