Hudson Data released ML Graph, a groundbreaking software framework that provides the ability to easily create and analyze graph structures from relational databases. ML Graph provides machine learning capability on vast amounts of data previously beyond the scope of traditional modeling tools as it allows users to efficiently make queries on a collection of nodes and edges. “Analysis of relationships is a much more effective way to model future behavior,” said Hudson Data CEO, Menish Gupta. He added, “ML Graph allows individuals to model emergent behaviors and create sophisticated graph models that will provide better insights than previously possible using traditional ML techniques.”
ML Graph is suited for datasets that have a limited number of records available for a given object/node where supervised learning techniques cannot be effectively employed. The solution is ideal for creating large, undirected, unweighted, graph clusters as it applies the mathematical principles of graph theory, where a graph structure represents pairwise relationships between objects and entities. These objects correspond to mathematical abstractions called vertices (nodes) and each of the related pairs creates an edge
The core ML Graph library is optimized for maximum performance and compact graph representation. It easily scales to massive networks with hundreds of millions of nodes, and billions of edges. The clusters created can then be analyzed to gain insights and predict future events. “This software is a game changer for data analytics organizations,” said Rick Arturo, VP of Business Development. “With our RESTful API and SQL toolkit, graph structure analytics can easily be employed throughout an organization; that’s a very powerful tool.” ML Graph will be used for solving mission critical problems in Fraud Analytics, Financial Services, Insurance, Healthcare, and Logistics. For more information contact firstname.lastname@example.org.