Machine Learning And Business Problem-Solving

David Cruickshank

2018 at the SAP Co-Innovation Lab in Silicon Valley is off to a roaring start, with three of its projects heavily focused on the topic of artificial intelligence. This project work spans machine learning, cognitive computing, computer vision, and even touches the realms of natural-language processing.

For our lab, we began digging into the application of machine learning beginning in 2014, exploring its application in everything from supply chain optimization to factory automation and retail, including predicting terrorist attacks. Where we can apply knowledge for a given domain and weave it into a learning algorithm for the sake of doing non-deterministic pattern recognition, machine learning grounded in only statistics (not symbology, logic, or evolutionary) can readily improve upon guessing. Learning from a productive data set, and where overfitting is sufficiently avoided or mitigated, a learning algorithm can recognize patterns and generalize to cases not yet encountered. Such explorations for us started more than two years ago with SAP NS2 and ConvergentAI (formerly AxxonAI) where we find the project team’s proof-of-concept (POC) results remain relevant today, but applicable to problem-solving the same way in other domains.

While conceptually different, a strong relationship exists between machine learning and analytics where machine learning uses data and learning algorithms (supervised and unsupervised) to optimize a model based on performance and prior experience. The resulting model is often used to improve the accuracy of analytics. Predictive analytics uses this learned model to find patterns against new data used to make informed predictions about future events.

Applying swarm intelligence for a POC featuring event risk forecasting

In the co-innovation lab, the project team worked to integrate what ConvergentAI developed as a unique predictive analytics capability, combining a decentralized and continuous machine learning engine with a decentralized and continuous analytic forecasting model. ConvergentAI calls its predictive analytics “swarm intelligence.”

SAP NS2 is a part of the family of SAP companies. One area of focus is working with U.S. national security agencies. In its collaborative project work both within SAP and with select partners from its ecosystem, SAP NS2 brought a data fusion platform together in a way that cleverly incorporates the machine learning and analytics capabilities of ConvergentAI to provide an event risk forecasting capability.

This application of event risk forecasting is designed to reduce uncertainty or risk with a process model that connects past events to their possible origins to analyze and predict future events. Its general reasoning and learning approach can be applied to a wide range of domains.

Key to this integrated capability is the ability to ingest, exploit, make sense of, and create a knowledge base of information that can be used as the source for event risk forecasting. Event risk forecasting requires conditioned information in order to drive the swarming model to forecast future events.

SAP NS2 developed its data fusion architecture on SAP HANA as a proof of concept to support a general-purpose, closed-loop process to integrate and correlate data, and thus create a knowledge base within an in-memory database for exploration and discovery of information. The database also serves as the source for other applications such as the event risk forecasting engine developed by ConvergentAI.

Continuously updated risk forecasting

The swarm intelligence technology and corresponding models become meaningful to a human analyst in explaining why a particular output is generated. The event risk forecasting engine runs continuously, enabling such changes to be translated into changed risk forecasts, building the human operator’s intuition of the correspondence of domain assumptions to risk distributions. In contrast to many other machine learning applications, the event risk forecasting engine continuously reevaluates its assertions and conclusions based on currently available data and parameters.

Broad industry application

The co-innovation lab project fully demonstrated intelligent data fusion and event risk forecasting in an effort to predict future terror activity or a given actor’s relationship to an act of terror using a substantial data set of event violence in Africa. From this demo, it is easy to see how this capability can be applied domestically in law enforcement (smarter allocation of police), and also commercially for the protection of large-scale critical infrastructure (oil and gas operations, airfields, or other transportation). Beyond forecasting the probability of events with the intent to suppress them, industries such as insurance have an interest in predicting the risk for contingency planning purposes.

For a more complete understanding of our work on event risk forecasting at SAP Co-Innovation Lab, we invite you take a look at our white paper “Predictive Analytics with Intelligent Data Fusion Event Risk Forecasting.” To learn more or to possibly explore how this can be applied to your business, please comment and/or connect with us to explore this in richer detail.


David Cruickshank

About David Cruickshank

David Cruickshank is senior director for strategy and operations for the SAP Co-Innovation Lab. He leads the lab's efforts in Silicon Valley to enable ecosystem-driven co-innovation between SAP, its partners, and customers. Additionally, he manages all operational aspects necessary to run a multimillion-dollar data center to provision private cloud infrastructures to deliver productive SAP landscapes consumed by co-innovation projects seeking a faster track to market for commercially successful innovations.