Global Infrastructure API

Building an intelligently linked global infrastructure database with NLP and knowledge graphs

Problem

More financial institutions are investing in core-plus, value-added and opportunistic infrastructure investments*, but opaque markets and niche use cases mean that they struggle to gain deep insights across active and potential investment opportunities.

With intense and increasing competition for viable assets, investors need more timely and accurate market intelligence.

*Source: 2020 Pregin Global Infrastructure Report

Solution

LSEG Labs collaborated closely with our investment management and sovereign wealth fund clients to devise a novel method to wrangle, link and enrich six Refinitiv datasets related to infrastructure.

LSEG Labs uncovered 30% more relationships across infrastructure assets by creating an implicit linking method using knowledge graphs and natural language processing (NLP) to identify and link high-priority tables.

The project used named entity recognition on unstructured textual data as well as Refinitiv’s open, permanent, and universal identifiers for data, PermID, to pull in associated metadata into a knowledge graph and create a richer, context heavy dataset for analysis.

Global Infrastructure API in action

LSEG Labs built a systematic, qualified understanding of all infrastructure-adjacent coverage via projects, finances, participants, deals, and project details.

The prototype includes:

  • Coverage spanning 60,000 projects, 24,000 entities with PermIDs, 30,000 connected deals and 4M+ relations
  • LSEG Labs implicit linking method, using NLP and Refinitiv PermID, identified 550% more projects and entities and uncovering 30% more relationships
  • Six linked databases, wrapped in an easily queried RESTful API

What we’re thinking next

The Infra 360 API’s intelligently and richly linked data is now helping teams across LSEG create new platforms and insights into relatively obscure infrastructure markets.