Decision Management Community

Declarative AI is the leading international conference on the blending of rules, reasoning, AI, decisions, rule-based machine learning and explanations. It brings together:

• The 5th International Joint Conference on Rules and Reasoning (RuleML+RR 2021)
• The 8th DecisionCAMP 2021
• the 17th Reasoning Web Summer School (RW 2021)

Its program has many interesting papers directly related to BR&DM, e.g.:

• Learning decision rules or learning decision models?
• Deep Learning for the Identification of Decision Modelling Components
• Automatic Generation of Intelligent Chatbots from DMN Decision Models

Register

Colorectal cancer is a leading cause of death. Colonoscopy is the criterion standard for detection and removal of precancerous lesions and has been shown to reduce mortality. The polyp miss rate during colonoscopies is 22% to 28%. DEEP DEtection of Elusive Polyps (DEEP²) is a new polyp detection system based on deep learning, which alerts the operator in real-time to the presence and location of polyps. The primary outcome was the performance of DEEP² on the detection of elusive polyps. Link More

Initially Business Logic Component (BLoC) pattern was introduced by Google as a solution to handle states in Flutter applications. It allows you to reduce the workload on UI components by separating the business logic from them. Over time, other frameworks also started to support the BLoC pattern – see e.g. how to use the BLoC pattern with React. It brings many benefits of business logic externalization: Flexibility to update application logic, Reuse logic, Ease of testing, Scalability. It is only natural to implement the BLoC pattern using a rule engine and additionally move control over business logic from programmers to subject matter experts. Link

Linguistics for the Age of AI, a book by two scientists at Rensselaer Polytechnic Institute, discusses the shortcomings of current approaches to natural language understanding (NLU) and explores future pathways for developing intelligent agents that can interact with humans without causing frustration or making dumb mistakes. Current AI systems “are not humanlike: they do not know what they are doing and why, their approach to problem solving does not resemble a person’s, and they do not rely on models of the world, language, or agency,” they write. “Instead, they largely rely on applying generic machine learning algorithms to ever larger datasets, supported by the spectacular speed and storage capacity of modern computers.” Link

This article by Gadi Singer introduces “Three Levels of Knowledge — 3LK — namely instantaneous, standby, and retrieved external knowledge, due to factors of scale, cost, diversity of tasks, and continuous adaptation. Thrill-K will be introduced as the AI systems architecture blueprint that implements 3LK principles for next-generation AI.” Link

We highlighted Decision Intelligence. Compare with the 2018 trends. Link

Prof. Jan Vanthienen just reported that for the first time ever a decision table became a part of the government law in Belgium/Flanders. Link

Our Aug-2021 Challenge follows the popular Kaggle Titanic competition that uses Machine Learning to predicts which passengers survived the Titanic shipwreck. In our challenge we want you to use the Kuggle training and test data to create and test a special decision model called “Titanic Booking Service”. First, you should create a decision model with rules that produce one of the following advices for each passenger: 1) Bon Voyage; 2) Go at your own risk; 3) Don’t do it!

Decision logic is not software code. As a result, you can’t be satisfied with testing tools and techniques that software developers use. Testing decision requires a different approach, different techniques, and different tools. Granted, ensuring that your decision service can compile is useful. But, the end goal is really to ensure that your decision logic complies with your business objectives. On July 22  Carole-Ann Berlioz from Sparkling Logic will explore what makes decisions different, how to test them, and how to automate regression tests.  Link