Competitive clustering algorithms based on ultrametric properties

Fouchal, Said; Ahat, Murat; Amor, Soufian Ben; Lavallée, Ivan; Bui, Marc

doi:10.1016/j.jocs.2011.11.004

Cited by 13 publications

(7 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, in Ref. , a recursive clustering method was proposed to elect Cnodes, based on two clustering algorithms: LEACH, which is common clustering networks, and Fast and Flexible Clustering Algorithm (FFUCA Algorithm) Recursive_Leach (R‐Leach): LEACH uses the clustering technique that divides a network into subsets called clusters, where each cluster is formed of CH and nodes which belong to it.…”

Section: Numerical Resultsmentioning

confidence: 99%

Dynamic and adaptive detection method for flooding in wireless sensor networks

Mansouri

Mokdad

Ben‐Othman

et al. 2017

Int J Communication

View full text Add to dashboard Cite

Summary In recent years, wireless sensor networks (WSNs) have attracted an increasing attention in several fields. However, WSNs must be treated with significant challenges in their design due to their special characteristics such as limited energy, processing power, and data storage that make the energy consumption saving a real challenge. Also, regarding their distributed deployment in open radio frequency and lack of physical security, these networks are vulnerable and exposed to several attacks: passive eavesdropping, active attacks, and identity theft. In this paper, we propose a new method called accordion method to detect and apprehend denial of service attacks in WSNs. This approach is a dynamic and an adaptive method based on using clustering method which allows electing control nodes that analyze the traffic inside a cluster and send warnings to the cluster head whenever an abnormal behavior is suspected or detected. The proposed method relies on the analysis of the evolution of the threshold messages (alerts) sent in the cluster. The proposed method has been evaluated, and the obtained numerical results show its benefit compared with other detection methods.

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

Dynamic and adaptive detection method for flooding in wireless sensor networks

Mansouri

Mokdad

Ben‐Othman

et al. 2017

Int J Communication

View full text Add to dashboard Cite

show abstract

“…Another approach consists in a different clustering method. The number of cluster is not set and all the number of clusters are dynamically updated like a competitive agglomeration clustering [3] or [5].…”

Section: Discussionmentioning

confidence: 99%

Disease tracking service in urban areas

Cormier

Fouchal

Yahiaoui

2012

2012 8th International Wireless Communications and Mobile Computing Conference (IWCMC)

View full text Add to dashboard Cite

Tracking epidemic disease is a very challenging issue nowadays. The success of such process could help medical administration to stop diseases quicker than usual. In this paper, we suggest a methodology based on wireless sensor networks deployed over volunteers who agree to carry a light wireless sensor network. Sensors over the body will monitor some health parameters (temperature, pressure, ..) and will run some light classification algorithms to diagnosis first diseases on the volunteers and later on their neighbors. The classification methodologies used in this study are based on the SVM approach or on the Fuzzy C-Means one. Finally, the wireless sensor network will send aggregated data about the disease to some base stations which collect the results. The main contribution is to execute an on-line disease tracking program and to detect some information about how the disease is propagated.

show abstract

“…In order to remedy this effect, Murtagh defines an ultrametricity index which overcomes the chaining effect as the ratio between the number of triangles which are almost isosceles and the number of all triangles in the data [55]. Furthermore, a clustering based on ultrametric properties is described in [61]. The derived feature vectors serve as input for classification.…”

Section: Dimensionality Reduction (Dr) Vs Feature Selection (Fs)mentioning

confidence: 99%

Unsupervised Feature Selection Based on Ultrametricity and Sparse Training Data: A Case Study for the Classification of High-Dimensional Hyperspectral Data

2018

View full text Add to dashboard Cite

In this paper, we investigate the potential of unsupervised feature selection techniques for classification tasks, where only sparse training data are available. This is motivated by the fact that unsupervised feature selection techniques combine the advantages of standard dimensionality reduction techniques (which only rely on the given feature vectors and not on the corresponding labels) and supervised feature selection techniques (which retain a subset of the original set of features). Thus, feature selection becomes independent of the given classification task and, consequently, a subset of generally versatile features is retained. We present different techniques relying on the topology of the given sparse training data. Thereby, the topology is described with an ultrametricity index. For the latter, we take into account the Murtagh Ultrametricity Index (MUI) which is defined on the basis of triangles within the given data and the Topological Ultrametricity Index (TUI) which is defined on the basis of a specific graph structure. In a case study addressing the classification of high-dimensional hyperspectral data based on sparse training data, we demonstrate the performance of the proposed unsupervised feature selection techniques in comparison to standard dimensionality reduction and supervised feature selection techniques on four commonly used benchmark datasets. The achieved classification results reveal that involving supervised feature selection techniques leads to similar classification results as involving unsupervised feature selection techniques, while the latter perform feature selection independently from the given classification task and thus deliver generally versatile features.

show abstract

Competitive clustering algorithms based on ultrametric properties

Cited by 13 publications

References 8 publications

Dynamic and adaptive detection method for flooding in wireless sensor networks

Dynamic and adaptive detection method for flooding in wireless sensor networks

Disease tracking service in urban areas

Unsupervised Feature Selection Based on Ultrametricity and Sparse Training Data: A Case Study for the Classification of High-Dimensional Hyperspectral Data

Contact Info

Product

Resources

About