IDC predicts that our global datasphere—the digital data we create, capture, replicate, and consume—will grow from approximately 40 zettabytes of data in 2019 to a staggering 175 zettabytes in 2025.
As a result, it becomes obvious that with increasingly complex data landscapes, organizations can rapidly feel overwhelmed and not be able to deliver on the desired business outcomes. With the growing demand for self-service, traditional approaches for structuring, organizing, and modeling the data prior to using it for analytics have become too cumbersome and unable to adapt to the flexibility and agility demanded by users. With big data, we have seen the emergence of new approaches such as data lakes where data of all formats and types can be collected in the raw form.
While access to raw data has been simplified, this has pushed more work to the users analyzing the data. The lack of data and analytics skills have become the bottleneck. Data governance considerations are also further challenging the uses of data for analytics, forcing organizations to have more traceability and transparency in analytics.
Clearly neither the traditional approach nor the modern approaches of collecting raw data can scale to address the challenges that organizations will face moving forward. Instead users need to easily find, understand, evaluate, trust and analyze data, regardless of their proficiency level. This is what semantic graphs promise to deliver.
A semantic graph stores passive metadata describing the data in business terms, along with usage data about how often the data is accessed, by whom, and relationship data about where the data is coming from or how objects are used together in analysis scenarios. But a semantic graph also turns this passive metadata into active metadata. Active metadata is what allows queries and calculations over these business terms to be performed in a consistent and repeatable manner. But active metadata also extends traditional semantic layer capabilities by supporting new use cases, such as recommendations on what data to use, or guides the user by exposing a trust score for the data based on usage patterns.
A semantic graph captures, organizes, and enriches metadata in a graph representation and uses graph analytics techniques to derive insights. A semantic graph combines graph capabilities and metadata representation in order to provide an augmented semantic representation of the data landscape in support of analytical workflows. It is participative in nature: it leverages the work performed by the users when building analytics or simply visualizing data, and augments it by leveraging AI and ML techniques to contribute back to the community by providing trust and relevance scores, or discovering relationships with other data sets.
The semantic graph will become the backbone supporting data and analytics over a constantly changing data landscape. It will help overcome the challenges of lack of skills and increased data governance pressures by leveraging AI and ML, the knowledge of the community and providing transparency in the uses of the data. Organizations not using a semantic graph are at risk of seeing the ROI for analytics plummet due to growing complexity and resulting organizational costs.
See more of what today’s top thought leaders say requires your organization’s attention. Download 10 Enterprise Analytics Trends to Watch in 2020 to read insights from Constellation Research’s Ray Wang and Doug Henschen, Forrester analysts Srividya Sridharan, Mike Gualtieri, and J.P. Gownder, Ventana Research’s Mark Smith and David Menninger, IDC’s Chandana Gopal, Marcus Borba, and more. Learn more about MicroStrategy’s semantic graph capabilities today.