GRAKN.AI makes working with complex data easy by providing:
- an ontology that allows you to model the world and all the type hierarchies and hyper-relationships contained within it.
- a query language that uses machine reasoning for retrieving explicitly stored data and implicitly derived knowledge.
Expressive data model
Grakn allows you to model the real world.
Grakn allows you to model the real world and all the hierarchies and hyper-relationships contained within it. The ontology modelling constructs include but are not limited to data type hierarchy, relation type hierarchy, bi-directional relationships, multi-type relationships, N-ary relationships, relationships in relationships, conditional relationships, virtual relationships, dynamic relationships, and so on.
Logical integrity of information
Grakn maintains higher data quality with stronger information consistency than any other database.
Grakn’s ontology functions as a data schema constraint that guarantees the logical integrity of information stored in the database in real time. In consequence, data is maintained at a higher quality with much stronger information consistency than for any other database. Grakn allows your team to catch all data errors as soon as the data arrives in your system. In Grakn’s Enterprise solution, you can also have real-time alerts of violations of advanced business rules.
Flexible data model
Grakn allows your business model to evolve regularly.
Grakn’s ontology is flexible. You can add new data and relation types, update type names, delete unused types, all while storing and retrieving data to and from the database. This allows your business model to evolve regularly even when you have lots of data, building a lasting advantage as your business “learns”. In Grakn’s Enterprise solution, you can have a machine learning system to automatically adapt and grow your Grakn ontology.
Machine reasoning query language
Graql allows you to discover deeply hidden and implicit associations within the data.
Graql is a polymorphic, knowledge-oriented query language. It allows you to query many different types simultaneously through a common supertype, and it takes into account deep contextualisation of data. Graql uses machine reasoning to perform inference over data types, relation types, context disambiguation, implicit relationships and dynamic relationships. This allows you to discover deeply hidden and implicit association between data instances through short and concise statements.
Optimised query execution
Graql reduces the time and complexity in writing deep relational queries while automatically optimising the execution of query traversal.
When you’re writing queries to execute on highly interconnected data, you usually have to think hard about the structure of your query so that it executes the most optimal query traversal. However, because it is a declarative query language, Graql allows you define your query as a set of data patterns that could be written in any structure, and translate it to the most optimal query traversal for execution. This reduces the time and complexity in writing deep relational queries while automatically optimising the execution of query traversal.
Graph analytics out-of-the-box
Grakn allows you to spend more time analysing your data and less time coding algorithms for analysis.
Graql is also capable of performing distributed graph analytics as part of the language, which allows you to perform analytics over large graphs out of the box. These types of analytics are usually not possible without developing custom distributed graph algorithms that are unique to individual use cases. With Grakn, you can spend more time analysing your data, and less time coding the algorithm for analysis.