Today, SAP has announced three initiatives that expand and accelerate our machine learning capabilities.

First, we are launching SAP Leonardo Machine Learning, an exciting new offering that embeds machine learning into a new wave of applications. Second, we are opening our SAP Leonardo Machine Learning Foundation to the SAP ecosystem via SAP Cloud Platform. And third, SAP has joined the Partnership on AI, a broad collaboration of commercial and non-profit organizations to benefit people and society.

SAP Leonardo Machine Learning: Wave of New Applications to Power Every Part of Every Major Industry

SAP Leonardo Machine Learning is part of the new SAP Leonardo portfolio, which combines machine learning, Internet of Things (IoT), blockchain, analytics, Big Data, and data intelligence into a holistic digital innovation system allowing our customers to innovate at scale and redefine their business.

Discover the power of machine learning. It finds patterns, learns, and understands by recognizing voice, text, images, and videos, fueling business growth in ways never imagined possible. Realize its full potential with SAP Leonardo Machine Learning. Watch a video to learn more.

Today, we announce the following SAP Leonardo Machine Learning applications.

Enhancing the Customer Experience

SAP Service Ticketing: Automatically tag service tickets in the right category

SAP Service Ticketing (part of SAP Hybris Cloud for Customer) classifies incoming customer service tickets so that they can be routed to the right agent. The agent is then prompted with recommended solutions to improve operational efficiency. This is the first step towards self-running customer service. SAP Leonardo Machine Learning uses existing data – in this case the text from service tickets – and builds powerful new functionality from this valuable resource.

“Businesses thrive on data. With SAP Leonardo Machine Learning, BASF is able to derive learnings from our rich datasets, and to drive business process improvements that directly benefit our customers" said Wiebe van der Horst, Chief Information Officer, BASF Group.

SAP Customer Retention is another example of how SAP Leonardo Machine Learning translates data into actions and value. Many businesses maintain historical databases containing information about customer behavior, demographics, and information regarding whether a customer was lost to a competitor (“churned”). SAP Leonardo Machine Learning mines this data to discover the subtle but powerful indicators of churn, substantially improving your ability to predict it, and act in time before churn happens. SAP Customer Retention also forecasts cross- and up-sell, as well as loyalty, allowing businesses to choose the best actions to maximize the customer experience.

Automating Financial Services

SAP Cash Application (part of SAP S/4HANA Cloud) learns by observing how humans match incoming bank statements to open receivables like invoices. SAP Leonardo Machine Learning learns to perform this time-consuming and error-prone task automatically. This enables substantial reduction of labor cost per invoice and faster clearing, ultimately leading to improved process efficiency and service quality.

Supercharging Marketing

SAP Brand Impact: Capturing brand logo placement in near real time

Many companies promote their brands through placement of logos within sporting events and other televised content. Checking that brands appear as contractually committed by content providers has historically been done manually: taking four to six weeks to compile, and an entire quarter to adjust if needed. SAP Brand Impact uses deep learning to recognize brand images in videos and images in real time with an unmatched accuracy. Results are delivered immediately, providing a substantial advertising impact benefit.

“Audi’s sponsorship team found the SAP Brand Impact solution a very useful tool. It can help Audi to evaluate its sponsorship exposure at high levels of operational excellence and transparency", said Thomas Glas, Global Head of Audi Sports Marketing.

“We are impressed by the solution’s capabilities, precision and speed and see it as a great potential tool for our sales efforts, adding data and numbers to media exposure previously deemed unmeasurable", said Yaron Talpa, Chief Marketing Officer, Maccabi Tel Aviv Basketball Club.

Improving HR

SAP Resume Matching (part of SAP Fieldglass) focusses on recruiters. It starts with a job description, and scores candidates to match open positions based on an applicant’s experience and objectives as well as the ability to search a giant database of job openings and suggest potential recruits. It automatically extracts skills, qualifications, and profile information from resumes and job descriptions without the need to create a cumbersome list of synonyms. SAP Resume Matching allows recruiters to focus on the important part: interaction with candidates.

SAP Job Standardization (also part of SAP Fieldglass) guides recruiters and hiring managers to create accurate and unbiased job descriptions that are neither overly specific nor use “creative” job titles. Standardized job descriptions make it easier for recruiters to find the right candidate, and for candidates to find the best job.

Enhancing Productivity

To maximize the benefit of machine learning technology, we are committed to making enterprise applications easy to use through conversational applications. SAP CoPilot uses machine learning to simplify how people interact with computers, supporting this collaboration as an automated digital assistant. Users chat with SAP CoPilot using human-like natural language. And CoPilot provides a single-point-of-contact for all SAP applications as well as other vendor’s business solutions like Slack or Google G Suite. SAP is developing chatbots for conversational HR, procurement, service and support on our own conversational engine as a first step toward comprehensive conversational enterprise applications.

Beyond applications: SAP Leonardo Machine Learning Foundation

SAP makes it incredibly easy to enable intelligent features in enterprise software by integrating the intelligence directly into our portfolio applications. But our approach goes far beyond: With the SAP Leonardo Machine Learning Foundation , we open our core technology so that our customers and partners can build their own new machine learning applications using intelligent services embedded in SAP Cloud Platform.

These services can use customers’ own training data, and/or can be combined with pre-existing business or technical APIs. This building-block approach brings agility, extensibility, and integration into SAP and non-SAP solutions alike, ensuring that the intelligent applications we build together remain at the leading edge, providing state-of-the-art value to customers and their organizations.

SAP Cloud Platform also provides access to the SAP HANA data platform, which enables rapid processing of large data sets for real-time machine learning analysis. SAP HANA offers over 90 algorithms in its predictive analytics library (PAL) and integration to the R programming language.

Why you should take machine learning seriously

The announcement of SAP Leonardo Machine Learning marks a historic moment: the dawn of a new age of machine intelligence.

For decades, SAP solutions have been at the “heart” of business worldwide. 76% of the world’s transaction revenue touches SAP systems. With SAP Leonardo Machine Learning, we add a computer “brain” and nervous system into the enterprise “organism.”

What this means for you and your company: this massive data resource is now a source for insights that allow you to improve revenues and efficiencies while driving customer and employee satisfaction.

What’s common to these applications: humans working hand-in-hand with machine intelligence to produce business value that is much greater than either one working alone. Furthermore, SAP Leonardo Machine Learning includes breakthrough technology that goes beyond insights to drive actions that lead to substantially better business outcomes.

Jointly Committed on the Intelligent Enterprise

SAP partners with the best technology providers in the market. Google’s track record in Artificial Intelligence (AI) research and in deploying large scale cloud solutions makes them a trusted partner for SAP. TensorFlow is the first-choice Machine Intelligence framework used by SAP for designing, building and executing large-scale machine learning algorithms. NVIDIA powers the training of our massive data sets and deep learning algorithms – the basis for SAP’s machine learning applications. SAP uses NVIDIA DGX-1, integrated hardware and software supercomputer, to build machine learning enterprise solutions for our 320,000 customers.

SAP Joins Partnership on AI

The Partnership on AI (PAI) has announced SAP’s membership in its first wave of new members: a wide-ranging group of leading global companies, institutions and organizations. The PAI studies and formulates best practices on Artificial Intelligence (AI) technologies, advances the public’s understanding of AI, and serves as an open platform for discussion and engagement about AI and its impact on people and society.

SAP has joined at the formative stage of this initiative, and will contribute with our expertise and experience.

Joining the Partnership on AI demonstrates SAP’s commitment to ensuring that our customers and partners will always use AI responsibly and ethically to benefit their organizations, employees, end customers, as well as society as a whole.

What’s Next?

Shipping the first SAP Leonardo Machine Learning applications is only the beginning. Today, teams across SAP are already building the next wave of intelligent solutions. Here are some examples:

• Supply chain: A recommender machine learning application will supercharge the supply chain, advising on new suppliers and new products to buyers, and detecting unusual buying patterns.
• Financial Services: Intelligent risk modeling for a marketplace that lets suppliers apply for receivables funding, allowing banks to maximize revenues while minimizing the risk of default.
• Retail: A computer vision application to recognize a Stock Keeping Unit (SKU) from photos and videos of products on store shelves: helps to verify compliance with agreements about shelf placement, and tracks sales velocity.
• Invoice analysis: A system that automatically extracts remittance advice and payment information from invoices to automate the clearing process.
• Accounts payable: A system that automatically extracts information from invoices and matches to the right vendor and converts unstructured invoice line items to structured information.