Clark, P. and Nibblet, T. (1989). The cn2 induction algorithm. Machine Learning, 3(4):261--283. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
First 50 documents Next 50
Rule Induction for Classification of Gene Expression Array.. - Liden, Asker, Boström (Correct)
....various techniques such as divide andconquer (recursive partitioning) that generates hierarchically organized rules (decision trees) 9] and separate and conquer (covering) that generates overlapping rules. These may either be treated as ordered (decision lists) 10] or unordered rule sets [11]. Common for all these methods is that they are very attractive with regard to the analysis of input feature importance. Since the rule induction process in itself takes redundancy of input parameters into account, and that the process will seek to use the most significant features first, ....
Clark,P. and Niblett,T. (1989) The CN2 Induction Algorithm. Machine Learning, 3, 261283
Learning Patterns in Noisy Data: The AQ Approach - Michalski, Kaufman (2001) (Correct)
....other rule learning systems. To this end, we performed a series of experiments using different datasets. In the experiments, the Q(w) measure used with varying parameter w was compared with the information gain criterion (Section 3) the PROMISE method [1] 9] and the methods employed in the CN2 [5], IREP [8] and RIPPER [6] rule learning programs. To simplify the comparison, we use the uniform notation for all the methods. As was mentioned above, the information gain criterion takes into consideration the entropy of the examples covered by the rule and not covered by the rule, and the event ....
.... when P = N and p exceeds n (the latter presumably occurs in any rule of value in an evenly distributed domain) the PROMISE value reduces to: p n) P (9) To see this, note that when P = N, p P) N n) N) 1 can be transformed into (p P) P n) P) 1, which is equivalent to (9) CN2 [5] builds rules using a beam search, as does the AQ type learner, on which it was partially based. In selecting a rule, it attempts to minimize, in the case of two decision classes, the following expression: p (p n) log 2 (p (p n) n (p n) log 2 (n (p n) 10) This expression ....
[Article contains additional citation context not shown here]
Clark, P. and Niblett, T., "The CN2 Induction Algorithm," Machine Learning 3, pp. 261283, 1989.
Incremental Rule Learning with Partial Instance Memory for.. - Maloof (2003) (Correct)
....is an instance of the setcovering problem [16] with the added restriction that covers of one set cannot intersect with points of another. An NP hard problem, the AQ algorithm is a heuristic method for inducing rules from examples under this constraint [17] Other rule learning methods include CN2 [18], IREP [19] and RIPPER [20] The AQ algorithm begins by selecting a positive example and generalizing it without covering any negative example, thereby forming a rule. It then removes the positive examples the rule covers and repeats with the remaining positive examples until covering them all. ....
P. Clark and T. Niblett, "The CN2 induction algorithm," Machine Learning, vol. 3, pp. 261--284, 1989.
Comprehensibility and Overfitting Avoidance in Genetic.. - Becker, Seshadri (Correct)
....complexity or biasing toward simplicity may involve post pruning of models or generating and hypothesizing models in the order from simple to complex or searching in the space of solutions from general to specific and using some stopping criterion. There have been a number of studies ( 5] [6], 7] where accuracy has not been reduced or has even been improved as a result of simplifying trees by pruning. There have also been theoretical arguments in favor of what has sometimes been referred to as Occam s Razor, namely that simpler models have greater predictive power and lead to less ....
Clark, P. & Nibblet, T. 1989. The CN2 Induction Algorithm, Machine Learning 3:261-283.
Toward Multidatabase Mining: Identifying Relevant Databases - Liu, Lu, Yao (2001) (Correct)
....user s interest, prior knowledge, and intention can be conveyed by means of queries. 2) Among various data mining tasks (classification, characterization, generalization, association, etc. most of them aim at discovering knowledge that can be expressed as relationships among values of attributes [4], 23] 11] 27] 30] 8] For example, classification rules, characteristic rules, and association rules can be expressed in the form of A B, where both A and B are conjunctions of attribute values. Queries can easily take care of these forms of knowledge. There are two key issues ....
....40 percent support and 100 percent confidence. We do not think such information is really interesting since on diet = no in fact holds for all tuples. In other words, what we found is not specific to Chinese. We can also apply certain classification algorithms, such as CART, CN2, and C4.5 [3] [4], 27] to each database, specifying attribute group as the class label. Using C4.5, for example, we can induce classification rules from both databases whose accuracy rates are better than that of just choosing the most probable class label. Therefore, based on whether any rules can be induced, ....
P. Clark and T. Niblett, "The CN2 Induction Algorithm," Machine Learning, vol. 3, pp. 261-283, 1989.
Gene Selection By Sequential Search Wrapper.. - Inza, Sierra.. (2002) (Correct)
....tree. The process stops at each node of the tree when all cases in that point of the tree belong to the same category or the best split of the node does not surpass a xed chi square signi cancy threshold. Then, the tree is simpli ed by a pruning mechanism to avoid overspecialization. The CN2 [8] algorithm represents a classi cation model by a set of IF THEN rules, where the THEN part represents the class predicted for the samples that match the conditions of the IF part. It is run with the default values of its parameters. CN2 is based on an information theoretic approach with a signi ....
P. Clark and T. Nibblet, `The CN2 induction algorithm', Machine Learning, 3(4), 261-283, (1989).
Experimental Comparison of Symbolic and Subsymbolic Learning - Wnek, Michalski (1992) (Correct)
....and decision tree algorithms AQ15, and ID3. The learned descriptions have either high R complexity (problem M2) or contain rules that cover noisy examples (problem M3) These problems were not learned as well by a hybrid of decision rules and decision trees, i.e. decision lists (CN2 algorithm, [29]) This suggests that techniques other then those implemented in these programs are required to solve this kind of problem. implemented in AQ17 HCI changes the representation space by narrowing and or expanding the initial set of attributes. The method analyzes inductive hypotheses generated by a ....
Clark, P. and Niblett, T. 1989. The CN2 Induction Algorithm, Machine Learning 3, pp. 261-284.
Knowledge Discovery in Texts Supporting e-Democracy - Paralic, Bednar (Correct)
....is the conjunction of the attribute tests (selectors) Rule induction algorithms have several advantages; the most noteworthy one is that these rules are best understandable for humans from all representations currently in use in concept learning. Examples of rule induction algorithms are CN2 [15] (see Fig. 2 with visualisation of the results in KDD Package [12] REP, IREP [16] RIPPER [17] Other possibility is to mine decision trees, e.g. using Quinlan s algorithm C4.5 Class. robo(s Accuracy. 100,00 Occunence 100,00 [21,31] Default. 27.87 Average. 100.00 Accuracy 100.00 Fig. 2: ....
Clark, P., Niblett, T. (1989) The CN2 induction algorithm. Machine Learning Journal, 3(4), 261-283.
Relational Learning - Clark And Thornton Self-citation (Clark) (Correct)
No context found.
Clark, P. and Niblett, T. (1989). The CN2 induction algorithm. Machine Learning, 3 (pp. 261-283).
Analysis of the Iterated Probabilistic Weighted k-Nearest .. - Martinez-Otzeta, Sierra (Correct)
No context found.
Clark, P. and Nibblet, T. (1989). The cn2 induction algorithm. Machine Learning, 3(4):261--283.
Evolutionary Learning of Hierarchical Decision Rules - Aguilar-Ruiz, Riquelme, Toro (2003) (1 citation) (Correct)
No context found.
P. Clark and T. Niblett, "The cn2 induction algorithm," Mach. Learn., vol. 3, no. 4, pp. 261--283, 1989.
Computational Intelligence Methods for Rule-Based Data.. - Duch, Setiono, Zurada (2004) (Correct)
No context found.
P. Clark and T. Niblett, "The CN2 induction algorithm," Mach. Learn., vol. 3, pp. 261--283, 1988.
Rule Evaluation Measures: A Unifying View - Nada Lavrac Peter (1999) (6 citations) (Correct)
No context found.
P. Clark and T. Niblett (1989). The CN2 induction algorithm. Machine Learning 3, pp. 261--284.
User Profiling in Personal . . . - Godoy, al. (2005) (Correct)
No context found.
Clark, P and Niblett, T, 1989, The CN2 induction algorithm. Machine Learning 3(4), 261--283.
An Analysis of Rule Learning Heuristics - Johannes Urnkranz Austrian (Correct)
No context found.
Clark, P., & Niblett, T. (1989). The CN2 induction algorithm. Machine Learning, 3, 261--283.
Decision Support Through Subgroup Discovery: - Three Case Studies (Correct)
No context found.
Clark, P. & Niblett, T. (1989). The CN2 Induction Algorithm. Machine Learning 3, 261--283. Cohen, W. W. (1995). Fast Effective Rule Induction. In A. Prieditis & S. Russell (Eds.), Proc. of the 12th International Conference on Machine Learning (pp. 115--123). Morgan Kaufmann.
Delegating Classifiers - Esar Ferri Cferri (2004) (2 citations) (Correct)
No context found.
Clark, P., & Niblett, T. (1989). The CN2 induction algorithm. Machine Learning, 3, 261--283.
An Extended Transformation Approach to - Inductive Logic Programming (Correct)
No context found.
CLARK,P.AND NIBLETT, T. 1989. The CN2 induction algorithm. Machine Learning 3, 4, 261-- 283.
An Incremental Method for Finding Multivariate Splits for.. - Utgoff, Brodley (1990) (16 citations) (Correct)
No context found.
Clark, P., & Niblett, T. (1989). The CN2 induction algorithm. Machine Learning, 3, 261-283.
Evolutionary Learning of Hierarchical Decision Rules - Aguilar-Ruiz, Riquelme, Toro (2003) (1 citation) (Correct)
No context found.
P. Clark and T. Niblett, "The cn2 induction algorithm," Mach. Learn., vol. 3, no. 4, pp. 261--283, 1989.
RIONA: A New Classification System Combining Rule Induction and .. - Gora, Wojna (2002) (Correct)
No context found.
Clark, P. and Niblett, T. (1989). The CN2 induction algorithm. Machine Learning, 3, pages 261--284.
Computational Intelligence Methods for Rule-Based Data.. - Duch, Setiono, Zurada (2004) (Correct)
No context found.
P. Clark and T. Niblett, "The CN2 induction algorithm," Mach. Learn., vol. 3, pp. 261--283, 1988.
Evolutionary Learning of Hierarchical Decision Rules - Aguilar-Ruiz, Riquelme (2003) (1 citation) (Correct)
No context found.
P. Clark and T. Niblett, "The cn2 induction algorithm," Mach. Learn., vol. 3, no. 4, pp. 261--283, 1989.
Knowledge Discovery from Sequential Data - Höppner (2003) (Correct)
No context found.
Clark, P. and Niblett, T. (1989). The CN2 induction algorithm. Machine Learning, 3:262--283.
The Development of the AQ20 Learning System - And Initial Experiments (Correct)
No context found.
Clark, P. and Niblett, T., "The CN2 Induction Algorithm," Machine Learning 3, pp. 261-283, 1989.
First 50 documents Next 50
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC