IBS Software developed a bilingual Named Entity Recognition (NER) system for cargo logistics using Amazon Bedrock. The system processes English and Japanese emails to extract 23 types of information, such as air waybill numbers and delivery instructions. By employing knowledge distillation from Amazon Nova Pro to Amazon Nova Lite via Amazon Bedrock, IBS Software achieved a 95.085 percent F1-Score accuracy while reducing operational costs by 14x. AI
IMPACT Demonstrates how cloud-based AI services can be leveraged for specialized industry tasks, potentially reducing costs and improving efficiency.
RANK_REASON This is a case study of a company using a cloud provider's AI service for a specific business problem, not a release from a frontier AI lab.
Read on AWS Machine Learning Blog →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
