Unsupervised learning of prototypes and attribute weights

Frigui, Hichem; Nasraoui, Olfa

doi:10.1016/j.patcog.2003.08.002

Cited by 208 publications

(116 citation statements)

References 10 publications

Supporting

Mentioning

105

Contrasting

Unclassified

Order By: Relevance

“…Thus, the total feature vector per chunk contains 16 * 2 * 12 = 384 features. Experimental results for FCM feature weighting-based VQ can be found in the literature [4], [5], [6], [7]. In this paper, results for FE feature weighting-based VQ are presented.…”

Section: B Speech Processingmentioning

confidence: 99%

“…. , M [5], [6], [7]. Weight values were estimated using either F CM -based estimation technique [4], [5] or F E-based estimation technique [6].…”

Section: Fuzzy Feature Weightingmentioning

confidence: 99%

“…Therefore a weighting estimation technique can be applied to any clustering techniques provided that these clustering techniques employ an objective function-based optimisation method. For example, an F CM feature weighting has been applied to F CM clustering as seen in [5] and to K-means clustering in [4] and [6].…”

Section: Fuzzy Feature Weightingmentioning

confidence: 99%

“…Moreover there are also two kinds of weights which are weight vector and weight matrix as mentioned above, so there are totally 2 * 3 * 2 = 12 combinations found. Four of these 12 combinations have been used in [4], [5] and [6]. The other eight combinations are proposed in this paper.…”

Section: Fuzzy Feature Weightingmentioning

confidence: 99%

See 3 more Smart Citations

Fuzzy feature weighting techniques for vector quantisation

Tran

Sharma

et al. 2010

International Conference on Fuzzy Systems

View full text Add to dashboard Cite

Abstract-Vector quantization (VQ) is a simple but effective modelling technique in pattern recognition. VQ employs a clustering technique to convert a feature vector set in to a cluster center set to model the feature vector set. Some clustering techniques have been applied to improve VQ. However VQ is not always effective because data features are treated equally although their importance may not be the same. Some automated feature weighting techniques have been proposed to overcome this drawback. This paper reviews those weighting techniques and proposes a general scheme for selecting any pair of clustering and feature weighting techniques to form a fuzzy feature weighting-based VQ modelling technique. Besides the current techniques, a number of new feature weightingbased VQ techniques is proposed and their evaluations are also presented.

show abstract

Section: B Speech Processingmentioning

confidence: 99%

“…. , M [5], [6], [7]. Weight values were estimated using either F CM -based estimation technique [4], [5] or F E-based estimation technique [6].…”

Section: Fuzzy Feature Weightingmentioning

confidence: 99%

Section: Fuzzy Feature Weightingmentioning

confidence: 99%

Section: Fuzzy Feature Weightingmentioning

confidence: 99%

See 2 more Smart Citations

Fuzzy feature weighting techniques for vector quantisation

Tran

Sharma

et al. 2010

International Conference on Fuzzy Systems

View full text Add to dashboard Cite

show abstract

“…As an improvement, evolutionary algorithms such as the genetic algorithm (GA) [9], ant colony optimization (ACO) [10], particle swarm optimization (PSO) [11] and features weighted fuzzy c-means (WFCM) [12] algorithms have been applied to tackle these problems. Many researchers are gradually promoting the use of the FCM algorithm in medical diagnosis with these optimal algorithms.…”

Section: Introductionmentioning

confidence: 99%

Multiple fuzzy c-means clustering algorithm in medical diagnosis

Duan

2015

THC

View full text Add to dashboard Cite

Abstract. BACKGROUND:In recent years, the use of the fuzzy c-means (FCM) clustering techniques in medical diagnosis has steadily increased, because of its effectiveness in recognizing systems in the medical database to help medical experts diagnosing diseases. However, its performance is highly dependent on the randomly initialized cluster centroids which may allow the diagnosis to be trapped into the problem of the local optimum. OBJECTIVE: This paper proposes a multiple fuzzy c-means (MFCM) algorithm for medical diagnosis. METHODS:The new method optimizes the initial optimizing cluster centers by comparing the Euclidean distance between patient data. Further, this paper assigns a set of weights to the features of a certain disease to equalize their difference influence as a substitute for data normalization. RESULTS: The performance of proposed MFCM algorithm was demonstrated through dividing complex primary headache data into Migraine, Tension-Type Headache (TTH), Trigeminal Autonomic Cephalalgias (TACs) and other primary headache disorders. In addition the superiority of MFCM algorithm was proven by comparing analytical results with other state-of-the-art clustering methods. CONCLUSIONS: This MFCM method has shown a new application in medical diagnosis.

show abstract

A generic approach to semantic video indexing using adaptive fusion of multimodal classifiers

Kim

Frigui

Fadeev

2008

Int J Imaging Syst Tech

Self Cite

View full text Add to dashboard Cite

We present a novel method for fusing the results of multiple semantic video indexing algorithms that use different types of feature descriptors and different classification methods. This method, called Context-Dependent Fusion (CDF), is motivated by the fact that the relative performance of different semantic indexing methods can vary significantly depending on the video type, context information, and the high-level concept of the video segment to be labeled. The training part of CDF has two main components: context extraction and algorithm fusion. In context extraction, the low-level multimodal descriptors extracted by the different classification algorithms are combined and used to partition the feature space into clusters of similar video shots, or contexts. The algorithm fusion component assigns aggregation weights to the individual classifiers within each context based on their relative performance in that context. Results on the TRECVID-2002 data collections show that the proposed method can identify meaningful and coherent clusters and that the performance of the different labeling algorithms can vary significantly across different clusters. Our initial experiments have indicated that the contextdependent fusion outperforms the individual algorithms and the global fusion of those algorithms. We also show that using standard multimodal descriptors and a simple k-NN classifier, the CDF approach provides results that are comparable to the state-of-the-art methods in semantic indexing.

show abstract

Unsupervised learning of prototypes and attribute weights

Cited by 208 publications

References 10 publications

Fuzzy feature weighting techniques for vector quantisation

Fuzzy feature weighting techniques for vector quantisation

Multiple fuzzy c-means clustering algorithm in medical diagnosis

A generic approach to semantic video indexing using adaptive fusion of multimodal classifiers

Contact Info

Product

Resources

About