Semantic Search in DDC CWICR Database

Business Case

Problem Statement

Construction cost estimation requires finding relevant work items from large databases. Traditional keyword search fails when:

• Users describe work in natural language
• Terminology varies across regions and languages
• Similar work items have different naming conventions

Solution

DDC CWICR database provides pre-computed embeddings (OpenAI text-embedding-3-large, 3072 dimensions) enabling semantic similarity search across 55,719 work items in 9 languages.

Business Value

• 90% faster work item lookup compared to manual search
• Multi-language support: Arabic, Chinese, German, English, Spanish, French, Hindi, Portuguese, Russian
• Higher accuracy by finding semantically similar items, not just keyword matches

Technical Implementation