Industry Classification System

3 posts · Aug 2024 – Jun 2025

What this trail covers

This post details the revamp of Ramp's industry classification system, moving from a RAG (Retrieval Augmented Generation) approach to a more robust and enriched system. This involved developing new models and data pipelines to improve the accuracy and depth of industry categorization for financial data. Specifically, it introduces the use of transaction embeddings to enhance retrieval accuracy, building upon the previous iteration of the industry classification system by leveraging AI to improve retrieval.

All Ramp trails

Jun 2025How Ramp Fixes Merchant Matches with AIThis post details how Ramp improved merchant matching accuracy by leveraging AI, specifically focusing on fixing incorrect matches. It describes the technical approach to identify and correct misclassified merchants, likely involving model retraining or fine-tuning, and potentially new data pipelines for error correction. This builds upon the existing industry classification system by addressing a specific failure mode (incorrect matches) and enhancing the system's practical application.
Jan 2025From RAG to Richness: How Ramp Revamped Industry ClassificationThis post details the revamp of Ramp's industry classification system, moving from a RAG (Retrieval Augmented Generation) approach to a more robust and enriched system. This involved developing new models and data pipelines to improve the accuracy and depth of industry categorization for financial data. Specifically, it introduces the use of transaction embeddings to enhance retrieval accuracy.
Aug 2024Improving Retrieval on Ramp with Transaction EmbeddingsThis post details the technical implementation of using transaction embeddings to improve retrieval accuracy within Ramp's industry classification system. It describes the process of generating these embeddings and how they are integrated into the RAG system to enhance the quality of industry categorization for financial data.
Advertisement