Some Approaches to Improve the Interpretability of Neuro-Fuzzy Classi ers
نویسندگان
چکیده
Neuro-fuzzy classi cation systems make it possible to obtain a suitable fuzzy classi er by learning from data. Nevertheless, in some cases the derived rule base is hard to interpret. In this paper we discuss some approaches to improve the interpretability of neuro-fuzzy classi cation systems. We present modi ed learning strategies to derive fuzzy classi cation rules from data, and some methods to simplify the found rule base to improve the interpretability of the resulting fuzzy system.
منابع مشابه
Improving Naive Bayes Classiiers Using Neuro-fuzzy Learning 1
Naive Bayes classi ers are a well-known and powerful type of classi ers that can easily be induced from a dataset of sample cases. However, the strong conditional independence and distribution assumptions underlying them can sometimes lead to poor classi cation performance. Another prominent type of classi ers are neuro-fuzzy classi cation systems, which derive (fuzzy) classi ers from data usin...
متن کاملApplying Boolean Transformations to Fuzzy Rule Bases
Neuro-fuzzy classi cation systems allow to derive fuzzy classi ers by learning from data. The obtained fuzzy rule bases are sometimes hard to interpret, even if the learning method uses constraints to ensure an appropriate fuzzy partitioning of the input domains. This paper describes an approach to build more expressive rules by performing boolean transformations during and after the learning p...
متن کاملPerformance of weighted radial basis function classifiers
This paper describes Weighted Radial Basis Functions, a neuro-fuzzy uni cation algorithm which mixes Perceptrons and Radial Basis Functions. The algorithm has been tested as a pattern classi er in practical applications. Its performance are compared against those of other neural classi ers. The proposed algorithm has performance comparable or better than other neural algorithms, although it can...
متن کاملLearning hybrid neuro-fuzzy classi(er models from data: to combine or not to combine?
To combine or not to combine? This very important question is examined in this paper in the context of a hybrid neuro-fuzzy pattern classi(er design process. A general fuzzy min–max neural network with its basic learning procedure is used within (ve di3erent algorithm-independent learning schemes. Various versions of cross-validation and resampling techniques, leading to generation of a single ...
متن کاملCombining GP operators with SA search to evolve fuzzy rule based classifiers
The genotype±phenotype encoding of fuzzy rule bases in GA, along with their corresponding crossover and mutation operators, can be used by other search schemes, improving the behavior of these last ones. As a practical consequence of this, a simulated annealing-based method for inducting both parameters and structure of a fuzzy classi®er has been developed. The adjacency operator in SA has been...
متن کامل