TechPrufer
Get started
Hugging Face
Hugging Face logoHugging Face · Feature trail

NLP Pipeline for Inbound Message Classification

2 posts · May 2022

What this trail covers

Sempre Health implemented an NLP pipeline to automatically classify and respond to inbound patient messages, leveraging Hugging Face's language models. This system has successfully handled over 20% of incoming messages, significantly improving business scalability and operational workflow by reducing the load on their patient operations team. The implementation involved refining labeling strategies, selecting appropriate models and architectures, and accelerating the development process.

All Hugging Face trails
  1. Efficient Table Pre-training without Real Data: An Introduction to TAPEXIntroduced TAPEX (Table Pre-training via Execution), a novel method for pre-training language models on tabular data without requiring real-world data. TAPEX synthesizes a pre-training corpus by sampling executable SQL queries and their execution results over tables, effectively learning a neural SQL executor. This approach significantly improves pre-training efficiency and achieves new state-of-the-art results on table question answering and fact verification benchmarks, outperforming previous methods like TAPAS and TaBERT.
  2. How Sempre Health is leveraging the Expert Acceleration Program to accelerate their ML roadmapThis post details how Sempre Health leveraged Hugging Face's Expert Acceleration Program to build an NLP pipeline for classifying and responding to inbound patient messages. It highlights the challenges of improving upon a rule-based system, the benefits of using Hugging Face language models, and the impact of the program in accelerating their ML roadmap. The collaboration provided guidance on labeling strategies, model selection, and neural network architecture, leading to a production system that handles over 20% of incoming messages.