1. Home
  2. Refinitiv Labs
  3. Projects - what we're working on
  4. Global Infrastructure API Project

Global infrastructure API

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

Refinitiv Labs’ Global Infrastructure API links project, entity, and financial data to answer the growing demand for the growing alternative asset: infrastructure.

Problem and opportunity

Infrastructure funds are a hot investment - returning 8.7% annualized on average in the 10 years to June 2019*, meeting or surpassing the S&P Infrastructure and S&P Oil & Gas Indices.

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

Not only this, the intense competition for viable assets to invest in creates a market for timely, accurate market intelligence. Refinitiv Labs spotted an opportunity to link and enrich Refinitiv’s infrastructure-adjacent data into a rich suite of infrastructure products. Using knowledge graphs, the Labs linked data on the Belt and Road Initiative (BRI) with all known infrastructure projects in MENA and the Americas. This data was then enriched with financial data from the project finance, loans, and bonds databases to support financial data scientists working to improve their firm’s infrastructure investing strategies.

*Source:  2020 Pregin Global Infrastructure Report

Solution        

After speaking with customers such as infrastructure investment managers and sovereign wealth funds about the data, the team devised a novel method to wrangle six siloed databases across the business. Refinitiv Labs narrowed down the key areas of focus with subject matter experts and created an implicit linking method using knowledge graphs and Natural language processing (NLP) to identify and link high-priority tables.

Refinitiv Labs uncovered 30% more relationships across infrastructure by linking entities intelligently across databases. Their solution leverages named entity recognition on unstructured textual data and uses Refinitiv PermID to pull in associated metadata into a knowledge graph create a richer, context heavy dataset for analysis.

Refinitiv Labs' implicit linking method

Learn-it-all Labs

Making sense of financial data with Knowledge Graphs

Join Refinitiv Labs for a behind-the-scenes look at how they have used knowledge graphs to link disparate financial datasets tailored to capitalise on the growing interest in infrastructure assets.

In this session, Refinitiv Labs provide an introduction to using knowledge graphs on financial data and share their recent experience of using knowledge graphs to stitch data together, eventually creating a robust API to deliver the intelligently linked data.

Global Infrastructure API in action

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

This prototype includes:

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

Key learnings for data scientists and engineers in finance

  • Graphs - as connected data - are particularly relevant in finance as they help construct understandings of regional and counterparty exposure that might not be immediately clear from more relational formats
  • Graphs enable finding patterns in connected data and are a good way to store structured attributes and unstructured information
  • Graph databases are easy to scale and maintain for data models that need updating regularly, though bidirectional time-aware graphs may pose issues if queries are not designed to suit connected data

Free datasets and notebooks

Free datasets and notebooks for data scientists, developers and quants

The data exploration tool was built by Refinitiv Labs and includes Refinitiv knowledge graph content with Jupyter notebooks.

Start exploring and using Refinitiv’s data

Call your local sales team

Americas
All countries (toll free): +1 800 427 7570
Brazil: +55 11 47009629
Argentina: +54 11 53546700
Chile: +56 2 24838932
Mexico: +52 55 80005740
Colombia: +57 1 4419404

Europe, Middle East, Africa
Europe: +442045302020
Africa: +27 11 775 3188
Russia and CIS: +7 495 775 1200
Middle East & North Africa: +971 4 453 6768

Asia Pacific (Sub-Regional)
Australia & Pacific Islands:
+612 8066 2494 
China mainland: +86 10 6627 1095
Hong Kong & Macau: +852 3077 5499
India, Bangladesh, Nepal, Maldives & Sri Lanka:
+91 22 6180 7525
Indonesia: +622150960350
Japan: +813 6743 6515
Korea: +822 3478 4303
Malaysia & Brunei: +603 7 724 0502
New Zealand: +64 9913 6203
Philippines: 180 089 094 050 (Globe) or
180 014 410 639 (PLDT)
Singapore and all non-listed ASEAN Countries:
+65 6415 5484
Taiwan: +886 2 7734 4677
Thailand & Laos: +662 844 9576

 

Help & Support

Already a customer?

Office locations

Contact Refinitiv near you