Mining Actionable Knowledge Using Reordering Based Diversified Actionable Decision Trees

Subramani, Sudha; Wang, Hua; Balasubramaniam, Sathiyabhama; Zhou, Rui; Ma, Jiangang; Zhang, Yanchun; Whittaker, Frank; Zhao, Yiqi; Rangarajan, Sarathkumar

doi:10.1007/978-3-319-48740-3_41

Cited by 12 publications

(9 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They usually explore a single counterfactual antecedent with a single objective function, differing mainly on the optimization procedure. Heuristic approaches typically require an already fitted decision tree or ensemble of trees and use greedy algorithms to find groups of samples on which actions can change the outcome (Yang et al, 2003(Yang et al, , 2007Subramani et al, 2016). Heuristics can also search on every positive path of a decision tree (i.e., changes that result in a swap of outcome), searching positive paths that change the ensemble's outcome with the lowest cost (Tolomei et al, 2017).…”

Section: Related Workmentioning

confidence: 99%

Mining Pareto-Optimal Counterfactual Antecedents With A Branch-And-Bound Model-Agnostic Algorithm

Raimundo

Nonato

Poco

2021

Preprint

View full text Add to dashboard Cite

Mining counterfactual antecedents became a valuable tool to discover knowledge and explain machine learning models. It consists of generating synthetic samples from an original sample to achieve the desired outcome in a machine learning model thus helping to understand the prediction. An insightful methodology would explore a broader set of counterfactual antecedents to reveal multiple possibilities while operating on any classifier. Thus, we create a tree-based search that requires monotonicity from the objective functions (a.k.a. cost functions); it allows pruning branches that will not improve the objective functions. Since monotonicity is only required for the objective function, this method can be used for any family of classifiers (e.g., linear models, neural networks, decision trees). However, additional classifier properties speed up the tree-search when it foresees branches that will not result in feasible actions. Moreover, the proposed optimization generates a diverse set of Pareto-optimal counterfactual antecedents by relying on multi-objective concepts. The results show an algorithm with working guarantees that enumerates a wide range of counterfactual antecedents. It helps the decision-maker understand the machine learning decision and finds alternatives to achieve the desired outcome. The user can inspect these multiple counterfactual antecedents to find the most suitable one and have a broader understanding of the prediction.

show abstract

Section: Related Workmentioning

confidence: 99%

Mining Pareto-Optimal Counterfactual Antecedents With A Branch-And-Bound Model-Agnostic Algorithm

Raimundo

Nonato

Poco

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…At the end of this review the application of decision tree induction methods to the action mining problem should be mentioned [68,2,56,67,33]. Works [68,33] depicted transductive action mining methods.…”

Section: Related Workmentioning

confidence: 99%

“…These papers describe decision tree induction methods and optimal recommendation discovery based on the induced trees. Articles [2,56,67] focus on the post-processing of decision trees to discover the optimal recommendation. Decision tree ensembles have also been applied to the action mining problem [14,57].…”

Section: Related Workmentioning

confidence: 99%

SCARI: Separate and Conquer Algorithm for Action Rules and Recommendations Induction

Sikora¹,

Matyszok²,

Wróbel³

2021

Preprint

View full text Add to dashboard Cite

This article describes an action rule induction algorithm based on a sequential covering approach. Two variants of the algorithm are presented. The algorithm allows the action rule induction from a source and a target decision class point of view. The application of rule quality measures enables the induction of action rules that meet various quality criteria. The article also presents a method for recommendation induction. The recommendations indicate the actions to be taken to move a given test example, representing the source class, to the target one. The recommendation method is based on a set of induced action rules. The experimental part of the article presents the results of the algorithm operation on sixteen data sets. As a result of the conducted research the Ac-Rules package was made available.

show abstract

“…The output of the tree algorithm is input to post-processing algorithms which generate the action based on probability estimation of the classification from all the trees and at the end combine most actionable patterns. An improvement over this method has been proposed by Subramani et al [29]. In which a new technique for the construction of Decision Trees in order to get accurate and robust actionable patterns has been described.…”

Section: Literature Reviewmentioning

confidence: 99%

Actionable Knowledge Discovery for Increasing Enterprise Profit, Using Domain Driven-Data Mining

Batra

Rehman

2019

IEEE Access

View full text Add to dashboard Cite

Actionable Knowledge Discovery approaches to extract the business and technical significant actions/patterns to support direct decision making. These actions suggest how to transform an object from an undesirable status to a desirable status by incurring less cost and high profit. This article aims to propose a work that generates actionable patterns efficiently. It reduces the search space and number of iterations for attribute value change during action generation. Performance of the proposed method is compared with Yang's method and OF-CEAMA on the basis of four parameters i.e. the total number of rules required for action generation, run time of the methods, the total number of generated actions, total net profit and time and space complexity. Experiments have been carried out on four datasets retrieved from UCI Machine learning repository. Experimental results show that the proposed work takes less time than the other two methods to extract actions for all datasets. Also, the number of rules required to generate actions are less than the other two methods. Results also suggest that a decrease in execution time does not compromise the information and proposed work generates the same actions and net profit. Moreover, the proposed work tries to transfer an object from undesired status to the desired status. INDEX TERMS Actions, actionable knowledge, cost matrix, CRM, domain knowledge, domain driven, Knowledge Discovery, net profit.

show abstract

Mining Actionable Knowledge Using Reordering Based Diversified Actionable Decision Trees

Cited by 12 publications

References 9 publications

Mining Pareto-Optimal Counterfactual Antecedents With A Branch-And-Bound Model-Agnostic Algorithm

Mining Pareto-Optimal Counterfactual Antecedents With A Branch-And-Bound Model-Agnostic Algorithm

SCARI: Separate and Conquer Algorithm for Action Rules and Recommendations Induction

Actionable Knowledge Discovery for Increasing Enterprise Profit, Using Domain Driven-Data Mining

Contact Info

Product

Resources

About