Domain Driven Data Mining (DDDM, D3M) [1,4], or Domain-Driven Actionable Knowledge Discovery (AKD) [2,5-7,11], aims at building the next-generation data mining and analytics methodologies, techniques and tools that can discover and deliver knowledge and intelligence for decision action-taking, i.e., actionable knowledge and actionable intelligence . AKD/DDDM is interdisciplinary that integrates data mining, machine learning, organizational behavior, user interaction, human-computer interaction, and decision science.
Actionable knowledge refers to knowledge that can directly inform decision-making actions or directly support decision-making.
Actionable intelligence [1,2] refers to unique and valuable understanding, thinking, insights and expertise that can enable significantly better and smarter planning, decision-making and outcomes.
AKD/DDDM aims to directly address real-life enterprise problems and decision demand by incorporating and synthesizing ubiquitous data, information, knowledge and intelligence related to the underlying systems, data, behaviors, problems and environment per objectives and requirements, discover and deliver actionable outcomes satisfying both technical significance and business needs, and enable direct decision-making actions for enterprise, innovation, business, society, economy, and decision.
Actionable knowledge [2,3] has been qualitatively and intensively studied in the social sciences. Its marriage with data analytics and machine learning is only a recent story. It is a reality that the so-called knowledge discovered from data following the classic data mining and machine learning frameworks often cannot enable meaningful decision-making actions. This shows significant gaps between research and decision, and between knowledge, power, and decision-actions, and imbalance between research innovation and business need. A paradigm shift from knowledge discovery from data to actionable knowledge discovery and delivery is essential for the new-generation artificial intelligence and the era of data science, which should be actionable. Actionable knowledge discovery and delivery (AKD) [1-5] needs to systematically involve and integrate problems, data, behaviors, environment, models, decisions, and optimization into the whole objectives, processes, and outcomes of AI and data science.
The research topics include but are not limited to the following areas:
- AKD/DDDM methodologies: analytical and learning thinking, paradigms, mechanisms, processes, and architectures to enable actionable knowledge discovery and delivery;
- Intelligence representation: behavior intelligence, data intelligence, domain intelligence, symbolic intelligence, situated intelligence, connectionist intelligence, nature-inspired intelligence, social intelligence, organizational intelligence, network intelligence, and human intelligence in AKD/DDDM processes and systems;
- Knowledge representation: developing methods and tools to represent domain knowledge, expertise and expert experience, prior knowledge, and expert feedback in AKD/DDDM algorithms and systems;
- Intelligence metasynthesis: developing methods and mechanisms to synthesize diverse intelligences including human intelligence, domain intelligence, data intelligence, behavior intelligence, machine intelligence, network intelligence, social intelligence, and organizational intelligence; building consensus, managing divergence and convergence between intelligences into AKD/DDDM processes and systems;
- AKD/DDDM algorithms: analytical and learning methods, mechanisms, models, algorithms, and tools to discover and deliver actionable knowledge, intelligence and decision-actions;
- AKD/DDDM evaluation: analytical and learning evaluation systems, measures, tools to quantify, compare, and visualize actionable knowledge, intelligence and decision-actions;
- AKD/DDDM decision support: infrastructures, human-machine interactions, user interfaces, etc. to transform discovered knowledge to intelligence and decision-actions, and to embed knowledge and intelligence into decision systems;
- AKD/DDDM systems and tools: analytical and learning systems, tools, services, and applications to conduct actionable knowledge discovery and delivery.
The following diagram shows a concept map of AKD/DDDM.
Figure 1. The conceptual map of domain-driven, actionable knowledge discovery .
| Longbing Cao, Philip S Yu, Chengqi Zhang and Yanchang Zhao. Domain Driven Data Mining, ISBN: 978-1-4419-5737-5, Springer, 2010. BibTeX|
| Longbing Cao. Data Science Thinking: The Next Scientific, Technological and Economic Revolution, ISBN: 978-3-319-95092-1, Springer International Publishing, 2018. Download the frontmatter and backmatter. BibTeX|
| Philip Yu, Chengqi Zhang, Graham William, Longbing Cao, Yanchang Zhao.
Proc. Of ACM SIGKDD Workshop on Domain Driven Data Mining (edited), ACM Press (978-1-59593-846-6), 2007
 Longbing Cao. Domain-driven data mining: challenges and prospects, IEEE Trans. on Knowledge and Data Engineering, 22(6): 755-769, 2010. BibTeX
 Longbing Cao. Actionable Knowledge Discovery and Delivery, WIREs Data Mining and Knowledge Discovery, 2(2): 149-163, 2012. BibTeX
 Longbing Cao. Combined Mining: Analyzing Object and Pattern Relations for Discovering and Constructing Complex but Actionable Patterns, WIREs Data Mining and Knowledge Discovery, 3(2): 140-155, 2013. BibTeX
 Longbing Cao, Yanchang Zhao, Huaifeng Zhang, Dan Luo, Chengqi Zhang. Flexible Frameworks for Actionable Knowledge Discovery, IEEE Trans. on Knowledge and Data Engineering, 22(9): 1299-1312, 2010. BibTeX
 Longbing Cao and Tony He. Developing actionable trading agents, Knowledge and Information Systems: An International Journal, 18(2): 183-198 (2009). BibTeX
 Longbing Cao. Introduction to Domain Driven Data Mining, in Data Mining for Business Applications (eds. Cao L, et al.), 3-10, 2008.
 Yanchang Zhao, Huaifeng Zhang, Longbing CaoChengqi Zhang. Combined Pattern Mining: from Learned Rules to Actionable Knowledge, LNCS 5360/2008, 393-403, 2008. BibTeX
 Longbing Cao, et al. Domain-Driven, Actionable Knowledge Discovery, IEEE Intelligent Systems, 22 (4):78-89, 2007. BibTeX
 Longbing Cao, Chengqi Zhang. The evolution of KDD: Towards domain-driven data mining. International Journal of Pattern Recognition and Artificial Intelligence, 21(4): 677-692, 2007. BibTeX
 Longbing Cao, Chengqi Zhang. Domain-driven actionable knowledge discovery in the real world, PAKDD2006, LNAI 3918, 821 – 830. BibTeX
Please also refer to the AKD/DDDM special interest group for more information about the relevant concepts, activities, publications, and opportunities.