Handwritten numeral recognition using self-organizing maps and fuzzy rules

Chi, Zheru; Wu, Jian; Yan, Hong

doi:10.1016/0031-3203(94)00085-z

Cited by 118 publications

(36 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, we also included the C4.5 decision tree learner (Quinlan, 1993) as a wellknown benchmark classifier and, moreover, added two fuzzy rule-based classifiers from the KEEL suite (Alcalá-Fernandez et al, 2008): The CHI algorithm is based on Chi et al (1995Chi et al ( , 1996 and uses rule weighing as proposed by Ishibuchi and Yamamoto (2005). 6 The SLAVE algorithm makes use of genetic algorithms to learn a fuzzy classifier Perez, 1999, 2001).…”

Section: Classification Accuracymentioning

confidence: 99%

“…They are not very flexible and suffer from the "curse of dimensionality" in the case of many input variables but may have advantages with respect to interpretability (Guillaume, 2001). A well-known representative of this kind of approach is the CHI algorithm that we also used in our experiments (Chi et al, 1995(Chi et al, , 1996. It proceeds from a fuzzy partition for each attribute and learns a rule for every grid cell.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

FURIA: an algorithm for unordered fuzzy rule induction

Hühn

Hüllermeier

2009

Data Min Knowl Disc

386

223

View full text Add to dashboard Cite

This paper introduces a novel fuzzy rule-based classification method called FURIA, which is short for Fuzzy Unordered Rule Induction Algorithm. FURIA extends the well-known RIPPER algorithm, a state-of-the-art rule learner, while preserving its advantages, such as simple and comprehensible rule sets. In addition, it includes a number of modifications and extensions. In particular, FURIA learns fuzzy rules instead of conventional rules and unordered rule sets instead of rule lists. Moreover, to deal with uncovered examples, it makes use of an efficient rule stretching method. Experimental results show that FURIA significantly outperforms the original RIPPER, as well as other classifiers such as C4.5, in terms of classification accuracy.

show abstract

Section: Classification Accuracymentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

FURIA: an algorithm for unordered fuzzy rule induction

Hühn

Hüllermeier

2009

Data Min Knowl Disc

386

223

View full text Add to dashboard Cite

show abstract

“…An extension of Wang & Mendel algorithm [20] for classification problems proposed by Chi et al [2], that generates a fuzzy rule for each example in the training set and does not carry out any feature selection process. 2.…”

Section: Experimental Studymentioning

confidence: 99%

A Genetic-Programming-Based Approach for the Learning of Compact Fuzzy Rule-Based Classification Systems

Berlanga

Jesús

Gacto

et al. 2006

Artificial Intelligence and Soft Computing – ICAISC 2006

View full text Add to dashboard Cite

Abstract. In the design of an interpretable fuzzy rule-based classification system (FRBCS) the precision as much as the simplicity of the extracted knowledge must be considered as objectives. In any inductive learning algorithm, when we deal with problems with a large number of features, the exponential growth of the fuzzy rule search space makes the learning process more difficult. Moreover it leads to an FRBCS with a rule base with a high cardinality. In this paper, we propose a genetic-programming-based method for the learning of an FRBCS, where disjunctive normal form (DNF) rules compete and cooperate among themselves in order to obtain an understandable and compact set of fuzzy rules, which presents a good classification performance with high dimensionality problems. This proposal uses a token competition mechanism to maintain the diversity of the population. The good results obtained with several classification problems support our proposal.

show abstract

“…In general, the features can be classified in two great groups according to their nature: global, they are based on the association of patterns examining the image of a global form, and structural, treat the image analyzing topological aspects (for example, holes, contour, skeleton, etc.). Within the first type, we have worked with the zoning feature [1,3,10], that consists of dividing the box of the digit by zones and extract a vector of features on the basis of the black percentage of pixels that has each zone.…”

Section: Phase One: Muitifont Classifiermentioning

confidence: 99%

A hierarchical classifier for multifont digits

Rodríguez

Muguerza

Navarro

et al. 1998

Advances in Pattern Recognition

View full text Add to dashboard Cite

In this paper, the automatic recognition of broken and blurred, multifont typewritten digits in forms will be addressed. The classification, which is based on the utilization of a global feature, is divided in two phases: first, a minimum distance method (1-NN) is applied to provide a global classification of the patterns in a form; second, the patterns in the form previously classified are used to validate, or reject and reclassify them, on the basis of the mean distance to the predefined classes. In this way, a classification accuracy rate of 99.42% has been achieved.

show abstract

Handwritten numeral recognition using self-organizing maps and fuzzy rules

Cited by 118 publications

References 8 publications

FURIA: an algorithm for unordered fuzzy rule induction

FURIA: an algorithm for unordered fuzzy rule induction

A Genetic-Programming-Based Approach for the Learning of Compact Fuzzy Rule-Based Classification Systems

A hierarchical classifier for multifont digits

Contact Info

Product

Resources

About