Decision CAMP 2013 Speakers & Talks
Aziz Boxwala, MD, PhD, FACMI
|
The Health eDecisions project has developed specifications for interoperable knowledge artifacts for use in healthcare. These artifacts specify, for different clinical scenarios, the logic and applicable patient care actions in a computable form. The artifacts can be implemented within healthcare applications, such as clinical decision support systems, either as-is or translated into the native format of the application. The presentation will provide the motivation for the Health eDecisions project, the specification, and its applications.
Talk on Wednesday in the Healthcare track |
Dr. Benjamin Grosof
|
Motivated by requirements from many domains, including policy-based decision-making, we present two case studies in:
Talk on Tuesday |
Brian Stucky
|
The Financial Services industry was one of the earliest adopters
of Business Rule and Process approaches across a variety of applications. Process-centric enterprises where business rules were prevalent found these techniques to be a perfect fit for their day-to-day operations. As the platforms enabling rule and process solutions evolved, the Financial Services industry went through dramatic changes culminating in the crisis of 2008. Now Decision Management – both as a methodology and a technology – has been viewed as both a culprit and a necessity moving forward in a new era of financial regulation and scrutiny. The presentation will discuss the current state of Decision Management and its role in the new world of Financial Services. Talk on Wednesday at 8:30am in the Financial Services track |
Carlos Serrano-Morales
|
Automated Decision Management, relying mostly on the concrete application of business rules, predictive modeling and optimization technologies, has established itself as a core component of large business applications. While these applications are largely configured and managed based on the knowledge elicited from experts and data, they also need to take into account the fact that a lot is not known in advance in terms of the outcome of the decisions they make.
This talk will focus on approaches to achieve adaptive decision management, and in particular the corresponding technology and architecture implications. Adaptive Decision Management complements and extends rules, analytics and optimization approaches with the capability to include feedback from the outcome of decisions into online and offline processes that evolve the configuration of the decisions to account from the lessons learned from the analysis of these outcomes. Talk on TuesdayTalk on MondayCTO PanelCarlos will participate in the CTO Panel representing Sparkling Logic
Talk on Tuesday |
Carole-Ann Matignon
|
How do you extract your business rules? Do they reside in the head of your experts? Do they hide in your data?
There are 2 camps leveraging different technologies: analytics and business rules. In this talk, we will talk about the strengths and weaknesses of these approaches, and when and how you should use them. With the explosion of data, there is a fantastic opportunity to improve decision making. While accessibility has been addresses, the velocity and variety of data is becoming a new challenge. The data scientist shortage will force us to look at analytics in a different way. Are analytics applicable to all and any problems? Those are some of the considerations we will ponder while looking at practical usage of analytics and business rules. Talk on Monday |
Dr. Charles Forgy
|
Recent years have seen major advances in the expressiveness of rule languages. By taking advantage of these new capabilities, developers can write rules that are both easier to understand and easier for the rule engines to process.
In this talk, I will show examples of rules written in the older style alongside equivalent rules written in the new style. Measurements made of a typical rule engine running these rules will be presented. The results will show that the new features can sometimes make huge differences in the efficiency of programs. While the examples will be in a specific language (OPSJ) these capabilities are in fact available in many languages today, so the techniques shown are quite general. Talk on Tuesday |
Chris Adzima
|
Talk on TuesdayeBay has over 105 million active members. These customers communicate with each other using our platform by sending over 2 million messages a day. In some cases this communication is used to drive business away from eBay and to other platforms and companies. We call this activity Grey Market member to member communications. Not only does this grey market activity hurt eBay's revenue, it leaves buyers without the protection that transacting through eBay provides.
With so many messages sent through the eBay message system each day, we needed quick analysis and turnaround to ensure we had a product in production as soon as possible. Using an agile business rules approach, I created rules that would detect and prevent these grey market transactions from being initiated. This agility significantly reduced our cost to implement and has provided an amazing ROI. My rule set has increased site Gross Merchandise Sold (GMS) by $200 million a year. With the success of this initial phase, I am now creating more rules to increase detection and provide more revenue for the company. Key take-away • Greater ROI by utilizing a small team to manage and create business rules • Using Business rules to prevent customer to customer communications that harm company revenue Building business rules based on keyword data mined from customer to customer communications |
Christian Middleton
|
Keeping consumers engaged in their efforts to acquire and maintain a healthy lifestyle is a continuing challenge. In this talk, we describe a rule-based “eco system” in which information from a wearable sensor is synchronized with a smartphone that, in turn, synchronizes with the cloud.
Specifically, movement and sleep data is retrieved from the sensor device via the phone; this data is then integrated with dietary and geolocation data stored on the phone, and then aggregated with other user’s data on the cloud. This talk describes how Jawbone implemented a cloud based RBS in order to notify users with insights that inform them of their progress towards a healthy lifestyle. The social aspect of the application allows building rules that incorporate data from friends and people with similar characteristics, providing more interesting insights. Talk on Wednesday in the Healthcare track |
Colleen McClintock
|
Learn the basic concepts and benefits of decision management and get hands-on experience using SMARTS to develop a small decision
service! Please bring a laptop for hands-on exercises Wifi connection will be provided Talk on Monday in the Decision (boot) CAMP seriesLearn how to use SMARTS decision management environment
for decision services deployed to DROOLS. Talk on Wednesday in the Decision (boot) CAMP series |
Claude Nanjo
|
Today, Evidence-Based Clinical Decision Support knowledge exists primarily in document form and is generally population-based. However, there has been a recognition in the field that CDS knowledge needs to be patient-centric, dynamic, and computable in order to be more impactful at the point-of-care and within clinical workflows. To achieve this aim, CDS knowledge must be captured and structured on sound semantic models which are commonly understood and accepted.
Though no one model entirely meets this need today, this presentation will describe efforts in this direction thanks to encouraging trends in Industry. Talk on Wednesday in the Healthcare track |
Dr. Davide Sottara, PhD
|
Medicine is not an exact science and clinical practice is mostly based on evidence, which necessarily includes some degree of uncertainty. Clinical Decision Support (CDS), however, is traditionally based on (clinical) guidelines and rules, and implemented with a combination of (business) rules, workflow-based processes and event processing techniques.
None of these frameworks are suitable to model and process the uncertainty required by clinical decision making, while more suitable probabilistic and fuzzy techniques have not yet become mainstream technologies in the field. In this talk, we will discuss possible methodologies to integrate "hard" and "soft" decision making, ranging from loosely coupled systems to tightly coupled reasoners. Talk on Wednesday in the Healthcare track |
Dr. Emory Fry, MD
|
Dr. Fry will discuss the role of information technology in enabling
organizations to create cognitive environments conducive to the delivery of quality healthcare, highlighting synergies that can arise when workflow, decision-making, and resources are optimally aligned. He will introduce Socratic Grid, an emerging open source Clinical Decision Support (CDS) platform providing a living laboratory for exploring the knowledge management, business intelligence and predictive analytic technologies required for advanced cognitive and workflow support. Dr. Fry will argue that open-source, standards-based knowledge management architectures that can be integrated with diverse health information systems are essential to improving care and delivering quality healthcare at scale. Talk on Wednesday in the Healthcare track |
Dr. Evgeny Selensky
|
Businesses today face increasing challenges in trying to efficiently use available resources in order to maximize return on investment and minimize environmental impact. In the field scheduling and mobile resource allocation market, there are plenty of sophisticated scheduling products available. However, to start using these products business users have to feed them with such key characteristics of their mobile workforce as service locations where technicians can operate as well as their expertise (skills) in performing particular jobs such as equipment repair, testing, installation and maintenance. While an initial configuration of the field scheduling product is time-consuming, the real-world challenge is to maintain and adjust this information on a daily basis. So, to use these products efficiently, it is critical to automate these processes and to teach the scheduling system how to learn this information based on the actual field service assignments.
In this presentation we describe a rules-based mobile resource learner that addresses these problems in conjunction with a highly popular field scheduling tool. The learner enables the system to learn and adjust information about multi-level technical skills and geographic areas of the customer field service workforce. It allows a new customer to start using the scheduler with zero-configuration by just analyzing an actual history of technician work assignments. The learner uses a relatively small set of easily configurable rules that encode how workforce skills and locations are learned dynamically. The learned information is then fed into the scheduler/optimizer tool to improve schedule quality. The usage of the learner is demonstrated on a few simple examples. Talk on Tuesday |
Gil Segal
|
Today’s rules management systems and processes are chained to the
execution environment and lack the ability to rapidly adapt and evolve along with the ever-changing demands of the business. Sophisticated business logic management is critical to the organization’s success, yet few organizations have the ability to run the same business logic organization-wide, with consistency and accuracy. That is where decision management (DM) comes in. In this session, Gil Segal of Sapiens will share insights about the future of decision management and how decision management systems empower the business while freeing IT to do what IT does best. We will demonstrate how enterprise-scale decision management systems provide the ability to deploy the same business logic to multiple technological environments, and will discuss the benefits to IT of using a BDMS as part of an overall decision management architecture. Talk on Monday |
Dr. Harold Boley
|
Directed labeled graphs (DLGs) provide a good starting point for
visual data & knowledge representation but cannot straightforwardly represent nested structures, non-binary relationships, and relation descriptions. These advanced features require encoded constructs with auxiliary nodes and relationships, which also need to be kept separate from straightforward constructs. Therefore, various extensions of DLGs have been proposed for data & knowledge representation, including graph partitionings (possibly interfaced as complex nodes), n-ary relationships as directed labeled hyperarcs, and (hyper)arc labels used as nodes of other (hyper)arcs. Meanwhile, a lot of AI / Semantic Web research and development on ontologies & rules has gone into extended logics for knowledge representation such as object (frame) logics, description logics, general modal logics, and higher-order logics. The talk demonstrates how data & knowledge representation with graphs and logics can be reconciled. It proceeds from simple to extended graphs for logics needed in AI and the Semantic Web. Along with its visual introduction, each graph construct is mapped to its corresponding symbolic logic construct. These graph-logic extensions constitute a systematics defined by orthogonal axes, which has led to the Grailog 1.0 language as part of the Web-rule industry standard RuleML 1.0. Grailog has been used for teaching Logical Foundations of Cognitive Science and Semantic Web Techniques, and is being explored for PSOA RuleML querying of relational data in Clinical Intelligence. Talk on Wednesday in the Healthcare track |
Hidetoshi Sako
|
Insurance fraud is a growing problem in Asia and the majority of fraud detection systems have relied on statistics and analytics which require a sufficient quantity of fraud data to be effective. Small and medium-sized insurance companies, and many large insurance companies in Asia lack this data, and some attempts and deploying these systems have met with resistance from the business-side since they could not understand the output from these systems.
InsuPector is a next-generation insurance fraud detection system design to address these issues. Rather than relying solely on analytics and the availability of sufficient fraud data, InsuPector, built on Sparkling Logic SMARTS Decision Management System, combines analytics with over 500 fraud patterns that can be extended to meet the company's specific needs and refined and improved through a built-in performance monitoring feedback loop. InsuPector was designed to allow claims experts to customize and extend the business rules so that new products and fraud cases can be easily added. In this session you will learn:
Talk on Wednesday in the Financial Services track |
Jacob Feldman
|
Decision modeling practitioners frequently face situations when a decision model has to analyze multiple alternative decisions to come up with an optimal one. Traditional decision modeling approaches mainly rely on business rules and/or scorecards to determine one “good” decision and in most cases do not backtrack to consider potentially better decisions. The reason is simple: even for relatively small decision models with 20-40 decision variables it is physically impossible to manually create rules that cover all (!) possible combinations of decision variables and to find a decision that minimizes a total cost or maximizes resource utilization.
In this presentation we describe how “Decision Optimization” utilizes off-the-shelf optimization tools to help a business user to create smarter decision models that are capable to find decisions that minimize/maximize certain optimization objectives. Instead of trying to specify all possible rules, a decision model creator uses decision tables to represent only major business constraints and relationships between different decision variables. An optimization objective may be represented as a special decision variable defined on other key variables. A decision optimization component does the rest of work by automatically considering multiple alternatives and selecting the best one within a time limit defined by a user. Contrary to traditional rule engines, the optimization component is specifically designed to solve constraint satisfaction and optimization problems using proven optimization techniques such as constraint and/or linear programming. We will demonstrate decision optimization at work using two real-world use cases. Talk on TuesdayTalk on Monday |
James Owen
|
Talk on TuesdayMost initial BRMS (Business Rule Maintenance System) – or Rulebased – projects are often approached in the same manner: Someone hears of a successful project that has been done by another company, usually at a trade show or seminar, and they hear of the outrageous claims of savings in time and money and, being human, they want these same kind of savings for their company as well. So they contact a rule vendor or two or three and have them come to their company and do a presentation of what they can do for them. The vendor’s sales staff will, naturally, promise almost anything to get the order.
The customer, if they don’t have someone on staff who has been on such a project in the past, will be at the mercy of the vendor’s sales staff for the truth. Having been on upwards of 30 or more rule projects in many different industries over the past twenty years, mostly successful, some not, I have become somewhat experienced on what makes up a successful project. In this talk James will present some of the factors that make up a successful project and some of the pitfalls that a good manager should look out for to prevent a failure. Talk on Tuesday |
James Taylor
|
Decision Management is both an approach and a technology stack.
In this opening day workshop, Decision Management consultant and author James Taylor will introduce both. We'll begin with the discovery and modeling of suitable decisions, move into the construction of decision services and wrap up with the importance of decision analysis for continuous improvement. The critical technology capabilities - managing decision logic, embedding analytics, monitoring decision performance, and optimizing results - will all be introduced and presented in a coherent architecture for building Decision Management Systems. Different adoption paths and some best practices will conclude the session, putting you on a path to Decision Management success. Talk on Power Monday at 3pm in the Getting Started track |
Kenny Shi
|
Engineers like to build. Should we build a Business Rules Management System from scratch? What do you get from a vendor solution? What are the pain points at getting started on a vendor BRMS? What are the longer term needs adopting a vendor BRMS? What are the advantages of a home-grown BRMS? What are the biggest challenges building one in-house?
Having been there and done that, Kenny would like to share with audience the learning from building a BRMS, pros and cons from both sides. This talk will set the right expectations for those teams who want to build, and probably will give hints to vendors what capabilities are asked for by customers. Talk on Tuesday |
Diego Naya
|
The healthcare company has the largest coverage in Argentina, with over one and a half million affiliates. It is also the leader in healthcare technology usage of the region.
Over the last couple of years they have implemented a complex event processing system that takes advantage of all the cross references of different types of events. Along that time, they’ve compiled a series of best practices, design patterns and overall experience of its impact over operational aspects of the company. It is of their most interest to share this experience with other companies. This talk explains how to integrate a CEP server with enterprise level middleware, such as ESBs, legacy applications, different communication protocols such as SOAP, REST and JMS, and being able to process all events in a way that makes sense in a global way from the standing point of the company. This integration goes on to cover integrating outgoing events from the CEP server to the rest of the application world. We will cover as many best practices as possible, from defining and upgrading the event definitions to its transformations, from the data gathering to the information generation, and from coding to business ROI. Talk on Wednesday in the Healthcare track |
Mark Proctor
|
Writing classic computer games is a fun and challenging way to learn rule based programming.
This talk will develop games live, while explaining the mechanics of those things work. Starting with number guess and text adventures and eventually working up to Pong. Talk on TuesdayTalk on Monday |
Mike Weiss
|
The stock exchange space is an ever evolving world and to stay competitive one needs to have the ability to adapt quickly. Outside of the speed of an exchange’s trading platform, the next most important thing to attract business is a competitive pricing schema.
Firms will often determine where they are going to trade based on where they will get the best pricing. Because of this fact the business needs the ability to change rapidly to stay competitive. This talk will cover why Drools gives us an edge in this pricing war and how we are using it day to day. The talk will be broken into a couple major sections: • Why a rules engine? • Where we were before using business rules • Where we are currently at o Lessons learned to–date • Where we are doing with business rules Talk on Wednesday in the Financial Services track |
Neil Raden
|
Analytics is a broad subject area, but analytics in decision management has traditionally been a slow, artisanal process to create artifacts useful to rules engines at runtime. Those artifacts were created using techniques of data mining, predictive analysis and sometimes, optimization algorithms. But with the emergence of massively distributed processing grids, unstructured data and Hadoop, analytics has become something new - data science.
In this talk we'll discuss what the nature of data science and data scientists is, and apply some perspective on how it should and ultimately will affect your decision processes, both manual and automated. Great strides have been made informing decisions with analytical processes. However, the challenge of analytics is communication and creating a shared understanding. It’s about focusing on high impact areas, moving forward one step at a time, being skeptical, being creative, searching for the truth. Any company can compete on analytics. Talk on MondayNeil will moderate the CTO Panel:
Talk on Tuesday at 9:30am |
Ravi Singh
|
One of our industry’s biggest contention was the way business
requirements were captured and communicated between Business and IT. The Decision Model (TDM) was introduced by Barbara von Halle and Larry Goldberg in 2009 when they published their book The Decision Model - A Framework Linking Business and IT (Taylor&Francis, 2009). TDM is a new way to capture, organize, manage, test and author business logic unambiguously. TDM eliminated the guesswork for IT interpreting natural language statements submitted by the Business. This presentation will provide an overview of TDM and explain its profound impact on Business and IT. Talk on Tuesday |
Shash Hegde
|
The term "internet of things", loosely defined as a world where physical objects are seamlessly integrated into the information network, and where the physical objects can become active participants in business processes has gained popularity in recent years.
The talk will describe the case study of a small but visionary power management company in the Carolinas, which is building the next generation of a smart enterprise. An enterprise modeled on an event driven architecture which can sense and respond to stimuli faster than the competition. Achieving the vision has required our team to use a combination of data warehousing, decision management and analytics. Come learn about the vision, the challenges and also the business drivers for making this ambitious project a reality. Talk on Tuesday |
Srinivas Krovvidy
|
Fannie Mae has a long history of successfully deployed business rules applications since mid 1990s including launching of Desktop Underwriter(DU®), Custom Desktop Underwriter (Custom DU®),
Knowledge Acquisition and Rule Management Assistant (KARMA). However, the scope of these applications had been siloed and limited to specific business domains and the experiences and lessons learned from these applications did not translate into an enterprise vision until 2010. This presentation focuses on how an enterprise business rules management strategy was conceived in 2010 and various stages of its implementation during last three years resulting in the deployment of an enterprise business rules management system in 2011 and a centralized shared business rules service in 2012. It also highlights the ongoing activities including standing up a centralized rules governance process and adoption of various business applications that are planning to leverage the shared platform and business rules. The enterprise business rules platform enabled us to develop a streamlined license management process. We are also able to implement common business rules analysis, design and development best practices and templates across the organization. We are able to leverage skills and resources across multiple applications. We identified opportunities to share rules across applications resulting in consistent and cost effective policy automation. Talk on Wednesday in the Financial Services track |
