Empowering Data Mining Sciences by Habitual Domains Theory, Part II: Reaching Wonderful Solutions
- PDF / 1,667,260 Bytes
- 32 Pages / 439.37 x 666.142 pts Page_size
- 95 Downloads / 206 Views
Empowering Data Mining Sciences by Habitual Domains Theory, Part II: Reaching Wonderful Solutions Moussa Larbani1 · Po Lung Yu2,3 Received: 9 May 2020 / Accepted: 13 May 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract In the Part I of this paper, we presented the main concept of the proposed comprehensive decision model based on Habitual Domains theory, the concept of wonderful solution for solving challenging decision problems that we called decision making in changeable spaces problem (DMCS). In this Part II of the paper, we complete the construction of the model and show that it is operational and effectively empowers DMs in facing challenges. For this purpose, we present the mental principles “7–8–9 principles” that can be used to restructure decision parameters so that new solutions or alternatives could emerge. Then we provide procedures for finding wonderful solutions as sequences of the 7–8–9 principles by solving optimization in changeable spaces (OCS) problems, a new paradigm in optimization. Finally, we present applications of the model to post data mining analysis and decision making. In fact, the proposed model can be used in any area involving decision making and knowledge discovery such as management, politics, health care, technology and research. Keywords Post data mining · Habitual domain · Competence set · Wonderful solution
This paper is an extension of a keynote talk presented by the authors at the international conference on Information Technology and Information Management, ITQM 2018, Omaha, Nebraska, USA, October 21–22, 2018. * Moussa Larbani [email protected] Po Lung Yu [email protected] 1
School of Mathematics and Statistics, Carleton University, Ottawa, Canada
2
School of Business, University of Kansas, Lawrence, KS 66045, USA
3
Institute of Information Management, National Chiao Tung University, Hsinchu 30010, Taiwan
13
Vol.:(0123456789)
Annals of Data Science
1 Introduction In Sect. 3.4, Part I of the paper, we have mentioned ten factors (five decision elements and five decision environmental facets) that play a crucial role in real-world decision-making, including decision making in post data mining. Most of the existing decision models do not incorporate all the ten factors in a systematic way, especially, the psychological aspects, the external information input and the allowable time for solving a decision problem. Moreover, most of these models are generally Von Neumann-Morgenstern utility function based [9], and their solutions are computed as solutions of optimization problems when the objective function and constraints satisfy some analytical properties like continuity and convexity. These structural and conceptual constraints limit the application scope of utility functionbased decision models to challenging real-world problems such as post data mining decision-making (see Problems 2.1–2.4. Sect. 2, Part I of the paper). In the sequel, for ease of presentation, we will use “Part I” instead of “Part I of the paper” to refer to any par
Data Loading...
