DATA ANALYTICS

Empowering sales and marketing decisions through company knowledge graphs

 

Challenge identifier: DPC3-2017

 

Proposed by  

SpazioDati, an Italian consultancy developing a knowledge graph containing domain data across multiple vertical sectors combined with other information on multiple industries and euBusinessGraph, an innovation project co-funded by Horizon 2020.  Services include ontology development, data integration prototyping and assisting companies in discovering the value in their data.

Description

Company data is the foundation in the data value chain of the marketing and sales business sector. Marketing and sales managers make informed decisions based on official company data coming from authoritative data sources, such as company registers, but is increasingly influenced by supporting sources, such as Wikipedia, professional social networks and blogs.

SpazioDati is creating a large knowledge graph; a knowledge base containing interconnected data about companies in various European countries, including detailed descriptions of the relevant entities and events (for example, companies, subsidiaries, shareholders, business partners, managers, investments, mergers and acquisitions), as well as hidden relationships and patterns of corporate operations, within and across jurisdictions.  Understanding more about how entities operate, how entities are related to each other and what information exists about entities is vital to this process.

The challenge they propose is about enhancing this knowledge graph further, by addressing one or more of the following issues:

Company data contains hidden relationships and patterns that can provide valuable insights into companies. The challenge here is to discover this implicit information. A key question to answer is whether an examination of weblinks suggest that companies aggregate online based on their geographic location, by sector or size (of organisation). Another key question is that of capital concentration. Can it be shown that investors diversify by sector or geography when acquiring/controlling other companies or that they concentrate in similar companies?

It is often a challenge to link company data across different jurisdictions to establish influence.  This includes company linkages across countries via shared locations and managers.  How can the most influential companies/people be determined across sectors and countries? How might powerful holding companies best be identified? Across borders, how is it possible to assess which foreign investment is most influential on a national economy?

Finally, how can official company data be integrated with the Web data sources, such as Wikipedia. In particular, how can managers/shareholders be associated with their Wikipedia pages and and other subgraph extensions while also ensuring that any privacy rights arising can be respected?

Data

SpazioDati is providing access to parts of their knowledge graph containing information about companies from two EU countries, UK and Italy. The data includes:

  • Basic firmographics (e.g., name(s), incorporation data, registered addresses and ownership);
  • Financial data; We will not share balance sheets themselves, but ranges of important financial indicators, e.g., “revenue of a company falls into the range [0..10K] euros;
  • Companies directors and shareholders;
  • Locations (headquarters and sites);
  • Corporate websites and key entities extracted from them (such as products, technologies, services provided by companies).

More detailed information about the data can be found in our data catalogue. 

Expected outcomes

Algorithms that offer answers to the questions listed in the challenge, using SpazioDati’s data alongside public data sources such as Wikipedia, Wikidata etc. Applications need to explain how they will evaluate their algorithms to demonstrate that they produce useful results.

Expected impacts

  • Improved data value chain technology for corporate knowledge graphs.
  • More informed sales and marketing decisions based on richer, more accurate companies’ data

This project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under the Grant Agreement 732506. Unless otherwise indicated, all materials created by the Data Pitch consortium are licensed under CC-BY 4.0