Separate-and-Conquer Rule Learning

By Johannes Fürnkranz. 1999

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Interpretable Two-level Boolean Rule Learning for Classification (2015, methodology): “The second formulation replaces the 0-1 classiﬁcation error cost with the Hamming distance from the current rule to the closest rule that correctly classiﬁes a sample; we propose block coordinate descent and alternating minimization approaches for optimizing the objective in the second formulation.”
Semantic Web in data mining and knowledge discovery: A comprehensive survey (2016, background): “In their paper, they show that usingWordNet as a source of knowledge, and adapting a standard separate-and-conquer rule learning algorithm [162], they can significantly increase the semantic coherence in a rule without a decrease in classification accuracy.”
Explainable and Interpretable Models in Computer Vision and Machine Learning (2017, methodology): “For combining such rules into predictive theories, most algorithms follow the covering or separate-and-conquer strategy (Fürnkranz 1999), i.”
An experimental evaluation of simplicity in rule learning (2008, methodology): “(1) How does the predictive accuracy of SL(2) (as an algorithm that is designed to optimize for small rule sets) compare to the accuracy of state-of-the-art concept-learning algorithms? (2) How does the size of the rule sets that were induced by SL(2) compare to the size of those learned by state-of”
ENDER: a statistical framework for boosting decision rules (2010, methodology): “The most popular algorithms were based on a sequential covering procedure (Michalski 1983; Clark and Niblett 1989; Grzymala-Busse 1992; Cohen 1995; Fürnkranz 1996), also known as separate-and-conquer approach.”
Machine learning: a review of classification and combining techniques (2006, background): “Furnkranz (1999) provided an excellent overview of existing work in rule-based methods.”
Learning Complexity-Bounded Rule-Based Classifiers by Combining. Association Analysis and Genetic Algorithms (2005, background): “Well-known examples include the divide-and-conquer (recursive partitioning) strategy underlying decision tree induction [14] and the separate-and-conquer (covering) strategy that learns rules in succession, one at a time [6].”
pre: An R Package for Fitting Prediction Rule Ensembles (2017, methodology): “Following Frank and Witten (1998), we distinguish between two strategies for generating rules: Indirectly, through transforming the nodes of one or more decision trees to a set of rules, and directly, through for example a sequential covering approach (F¨urnkranz 1999).”
Advances in Intelligent Data Analysis V (2003, methodology): “An important conclusion of Bioch and Popova [5] is that with the increase of monotonicity noise in the data the number of conflicts criterion generates larger trees with a lower misclassification rate than the entropy splitting criterion.”
Exact Rule Learning via Boolean Compressed Sensing (2013, methodology): “The first rule set induction approach we consider, the set covering approach, also known as separate-andconquer, is the outer loop of myriad learning algorithms (Fürnkranz, 1999).”
Round Robin Classification (2002, methodology): “We chose Ripper (Cohen, 1995) as the base classiﬁer, which— in our view—is the most advanced algorithm of the family of separate-and-conquer (or covering) rule learning algorithms (F¨urnkranz, 1997; 1999b).”
Supervised Machine Learning: A Review of Classification Techniques (2007, unknown): “Furnkranz (1999) referred to most of them.”
FURIA: an algorithm for unordered fuzzy rule induction (2009, background): “RIPPER learns such rules in a greedy manner, following a separate-and-conquer strategy (Fürnkranz, 1999).”
Comprehensive Review of Artificial Neural Network Applications to Pattern Recognition (2019, background): “ANNs are excellent recognizers of patterns and robust classifiers [187]–[190] that have the capability to generalized when deciding on inaccurate data input [191]–[194].”
Directional decision lists (2015, methodology): “We end with an experiment on real data from a manufacturing process and determine indicators for problem cases.”
Comprehensible classification models: a position paper (2014, background): “Hence, some branch(es) may have to be added to the tree just to preserve the structure of a tree, even if that branch is associated with an irrelevant attribute value [9], [10], [24], [29].”
Probabilistic Inductive Logic Programming (2004, background): “Inductive logic programming systems that learn from entailment often employ a typical separate-and-conquer rule-learning strategy [13].”
A statistical approach to rule learning (2006, methodology): “When combined with suitable pruning heuristics, this procedure allows for eﬃcient and practically successful rule learning implementations (see (F¨urnkranz, 1999) for a survey).”