TY - JOUR
T1 - Behavior Action Mining
AU - Su, Peng
AU - Zeng, Daniel
AU - Zhao, Huimin
N1 - Funding Information:
This work was supported in part by the MOST Grant 2016QY02D0305, in part by the NNSFC Grant 71462001, in part by the NNSFC Innovative Team Grant 71621002, and in part by the CAS Grant ZDRW-XH-2017-3.
PY - 2019
Y1 - 2019
N2 - The actionable behavioral rules suggest specific actions that may influence certain behavior in the stakeholders' best interest. In mining such rules, it was assumed previously that all attributes are categorical while the numerical attributes have been discretized in advance. However, this assumption significantly reduces the solution space, and thus hinders the potential of mining algorithms, especially when the numerical attributes are prevalent. As the numerical data are ubiquitous in business applications, there is a crucial need for new mining methodologies that can better leverage such data. To meet this need, in this paper, we define a new data mining problem, named behavior action mining, as a problem of continuous variable optimization of expected utility for action. We then develop three approaches to solving this new problem, which uses regression as a technical basis. The experimental results based on a marketing dataset demonstrate the validity and superiority of our proposed approaches.
AB - The actionable behavioral rules suggest specific actions that may influence certain behavior in the stakeholders' best interest. In mining such rules, it was assumed previously that all attributes are categorical while the numerical attributes have been discretized in advance. However, this assumption significantly reduces the solution space, and thus hinders the potential of mining algorithms, especially when the numerical attributes are prevalent. As the numerical data are ubiquitous in business applications, there is a crucial need for new mining methodologies that can better leverage such data. To meet this need, in this paper, we define a new data mining problem, named behavior action mining, as a problem of continuous variable optimization of expected utility for action. We then develop three approaches to solving this new problem, which uses regression as a technical basis. The experimental results based on a marketing dataset demonstrate the validity and superiority of our proposed approaches.
KW - Business
KW - decision support
KW - knowledge and data engineering tools and techniques
KW - mining methods and algorithms
UR - http://www.scopus.com/inward/record.url?scp=85063805557&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85063805557&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2019.2896141
DO - 10.1109/ACCESS.2019.2896141
M3 - Article
AN - SCOPUS:85063805557
VL - 7
SP - 19954
EP - 19964
JO - IEEE Access
JF - IEEE Access
SN - 2169-3536
M1 - 8629898
ER -