Parameter estimation for Choquet fuzzy integral based on Takagi–Sugeno fuzzy model

Yue, Shihong; Li, Ping; Yin, Zong‐Xian

doi:10.1016/j.inffus.2004.11.002

Cited by 7 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To our knowledge, the best classification accuracy results have been obtained with IncNet (97.1%) [30]. With CI classifiers, the best reported classification accuracy is 91.5% [31]. As shown in Table 10, GI feature selection and CI classification yield a classification accuracy of 97.5%.…”

Section: Breast Cancer Datasetmentioning

confidence: 84%

Set characterization-selection towards classification based on interaction index

Murillo

Guillaume²,

Spetale

et al. 2015

Fuzzy Sets and Systems

View full text Add to dashboard Cite

In many real world datasets both the individual and coordinated action of features may be relevant for class identification. In this paper, a computational strategy for relevant feature selection based on the characterization of redundant or complementary features is proposed. The characterization is achieved using fuzzy measures and an interaction index computed from fuzzy measure coefficients. Fuzzy measure identification requires raw data to be turned into confidence degrees. This key step is carried out considering the distributions of feature values across all the classes. Fuzzy measure coefficients are then estimated with an improved version of the Heuristic Least Mean Squares algorithm that includes an efficient management of untouched coefficients. Then, a generalization of the Shapley index for an arbitrary number of features is used. Simulations experiments on synthetic datasets are performed to study the behavior of this generalized interaction index. For extreme datasets, containing either redundant or complementary features as well as noise, the index value is defined by mathematical formula. This result is used to motivate feature selection guidelines that take into account feature interactions. Experimental results on benchmark datasets show that the proposal allows for the design of compact, interpretable and competitive classification models.

show abstract

Section: Breast Cancer Datasetmentioning

confidence: 84%

Set characterization-selection towards classification based on interaction index

Murillo

Guillaume²,

Spetale

et al. 2015

Fuzzy Sets and Systems

View full text Add to dashboard Cite

show abstract

“…This implies that the students' performance in one subject does not have any relation with their performance in other subjects which is generally not true in many situations. However, Choquet integral plays an important role in capturing the interaction aspect among criteria during aggregation process [7]. Besides that, Choquet integral has the merit in producing unique solution in comparison to the other aggregation techniques [8].…”

Section: Choquet Integralmentioning

confidence: 99%

Choquet integral as an alternative aggregation method to measure the overall academic performance of primary school students: A case study

Kasim¹,

Abdullah²

2014

AIP Conference Proceedings

View full text Add to dashboard Cite

Abstract. Many average methods are available to aggregate a set of numbers to become single number. However these methods do not consider the interdependencies between the criteria of the related numbers. This paper is highlighting the Choquet Integral method as an alternative aggregation method where the interdependency estimates between the criteria are comprised in the aggregation process. The interdependency values can be estimated by using lambda fuzzy measure method. By considering the interdependencies or interaction between the criteria, the resulted aggregated values are more meaningful as compared to the ones obtained by normal average methods. The application of the Choquet Integral is illustrated in a case study of finding the overall academic achievement of year six pupils in a selected primary school in a northern state of Malaysia.

show abstract

“…Soon after, the corresponding theoretical investigations led to the increase of significance and practical relevance of non-additive set functions and nonlinear integrals [2][3][4][5][6][7][8][9][10][11]. While we fully acknowledge this state of affair, we remark that the concepts such as those pertaining to sequences of measurable functions analyzed in terms of Choquet integral deserve further attention, in particular in the context of their potential applications.…”

Section: Introductionmentioning

confidence: 98%

On the fundamental convergence in the (C) mean in problems of information fusion

Pedrycz

Yang

2009

Journal of Mathematical Analysis and Applications

View full text Add to dashboard Cite

The Choquet integral can be regarded as one of aggregation operators being used in information fusion. In this study, we offer an interpretation of sequences of measurable functions and the Choquet integral in the framework of information fusion. Based on an efficiency measure space, we also define a new concept of a fundamental convergence in the (C ) mean of sequences of measurable functions and discuss its theoretical underpinnings along with related interpretation issues as well as deliver some new results. Furthermore, an application of this concept is discussed in the context of information fusion. More specifically, based on the theoretical investigations, this idea is applied to the determination of a measurable function being used in the Choquet integral.

show abstract

Parameter estimation for Choquet fuzzy integral based on Takagi–Sugeno fuzzy model

Cited by 7 publications

References 23 publications

Set characterization-selection towards classification based on interaction index

Set characterization-selection towards classification based on interaction index

Choquet integral as an alternative aggregation method to measure the overall academic performance of primary school students: A case study

On the fundamental convergence in the (C) mean in problems of information fusion

Contact Info

Product

Resources

About