Our graph datasets can enhance existing tech solutions or be used to build novel solutions. We provide a way for tech teams to merge in-house, isolated data with the outside world quickly and access a variety of connected data analytical tools.

For example, credit and risk scoring providers can combine their data about individual companies with our network to provide network risk scores. No company exists in a vacuum and the behaviour and performance of directors and connected businesses will have a direct impact. That is valuable information which should be included.

Our demonstration app, Risk Hunter, uses only skeleton data from Companies House in a simplistic model (companies connect to officers, connect to companies, etc). Richer and more powerful models can be built quickly using our data pipelines and, for example, could incorporate:

– Shareholder/PSC data to show networks of ownership and influence and investment flows
– Addressing and other non-typical data points can be incorporated into models
– End client customer or supplier ledgers profiled through the prism of connected data can spot new opportunities and risk.

Connected datasets can be built from a combination of open, commercial or client data (and can include streamed data) and then modelled for general or specific use cases. Once the dataset is built it can then be queried with pattern, descriptive or predictive graph algorithms (from a library of over 50).

We have built our data assets on neo4j, the market-leading graph database, which provides unparalleled scaling, performance and data science capabilities.

Clients can consume via API or we can build a visual platform as per Risk Hunter’s UK company explorer.

For more information email mike.oaten@mnai.tech


For Neo4j Developers

We can provide read-only access to our full database for evaluation.

The key entities are UK ‘Companies’ (12m) and ‘CompanyOfficers’ (13m). There are properties for addresses, dates, company status. In all some 100m nodes and 500m relationships.

All we ask in return is for you to share your experience and findings with us. We are particularly interested in projects using graph algorithms and/or machine learning.

To request access email mike.oaten@mnai.tech