Ontotext offers GraphDB, an RDF database with built-in inference for the creation of knowledge graphs that link diverse data sources, and a platform that includes machine learning algorithms and APIs for building applications. “Ontotext can analyze textual content, enrich it with classifications according to a taxonomy, and recognize instances within a taxonomy,” said Ivaylo Kabakov, business unit head at Ontotext. “It works across heterogeneous data sources for entity recognition and can resolve ambiguity because the entity is referenced as a node that provides context.”
Based on their unique qualities, including their ability to easily adapt to new information and to provide deep insights about relationships in data among diverse repositories, graph technologies are expected to grow rapidly over the next few years. Gartner is predicting that the market will grow 28% per year, reaching $3.2 billion by 2025, and will be used in 80% of data and analytics innovations.
Enterprise Knowledge is a consulting company that often uses knowledge graphs to integrate data for their customers. “Knowledge graphs can aggregate information from disparate sources to cultivate a 360-degree view of core knowledge in an organization,” said Sara Duane, technical analyst for Enterprise Knowledge, “driving insights and connectivity.”
The way in which knowledge graphs are used depends on the industry. “In financial services, they are used to support categorization and connectivity in semantic layers,” Duane commented, “while pharmaceutical companies are leveraging them to drive research lineage and regulatory reporting. For organizations where content is a core asset, we’ve seen knowledge graphs transform the way that users receive timely and personalized content recommendations.”
The options are many and varied when it comes to taming silos. The specific approach should depend on the use case, the organization’s existing knowledge assets, and the strategic goals.