Search results for "Cluster Analysis"
showing 10 items of 848 documents
An approach based on the Adaptive Resonance Theory for analysing the viability of recommender systems in a citizen Web portal
2007
This paper proposes a methodology to optimise the future accuracy of a collaborative recommender application in a citizen Web portal. There are four stages namely, user modelling, benchmarking of clustering algorithms, prediction analysis and recommendation. The first stage is to develop analytical models of common characteristics of Web-user data. These artificial data sets are then used to evaluate the performance of clustering algorithms, in particular benchmarking the ART2 neural network with K-means clustering. Afterwards, it is evaluated the predictive accuracy of the clusters applied to a real-world data set derived from access logs to the citizen Web portal Infoville XXI (http://www…
A Semantic Collaborative Clustering Approach Based on Confusion Matrix
2019
In this paper we discuss about a new images retrieval technique based on clustering. We argue that images don’t have an intrinsic meaning, but they can receive different interpretation. These images can complicate documents retrieval. However, users need a quick and direct access to documents. To answer this requirement, we propose a retrieval approach which use a collaborative clustering technique based on Confusion matrix.
Combining fuzzy C-mean and normalized convolution for cloud detection in IR images
2009
An important task for the cloud monitoring in several frameworks is providing maps of the cloud coverage. In this paper we present a method to detect cloudy pixels for images taken from ground by an infra-red camera. The method is a three-steps algorithm mainly based on a Fuzzy C-Mean clustering, that works on a feature space derived from the original image and the output of the reconstructed image obtained via normalized convolution. Experiments, run on several infra-red images acquired under different conditions, show that the cloud maps returned are satisfactory. © 2009 Springer Berlin Heidelberg.
Heuristics for a Real-World Mail Delivery Problem
2011
We are solving a mail delivery problem by combining exact and heuristic methods. The problem is a tactical routing problem as routes for all postpersons have to be planned in advance for a period of several months. As for many other routing problems, the task is to construct a set of feasible routes serving each customer exactly once at minimum cost. Four different modes (car, moped, bicycle, and walking) are available, but not all customers are accessible by all modes. Thus, the problem is characterized by three interdependent decisions: the clustering of customers into districts, the choice of a mode for each district, and the routing of the postperson through its district. We present a t…
Data mining-based statistical analysis of biological data uncovers hidden significance: clustering Hashimoto’s thyroiditis patients based on the resp…
2014
The pathogenesis of Hashimoto's thyroiditis includes autoimmunity involving thyroid antigens, autoantibodies, and possibly cytokines. It is unclear what role plays Hsp60, but our recent data indicate that it may contribute to pathogenesis as an autoantigen. Its role in the induction of cytokine production, pro- or anti-inflammatory, was not elucidated, except that we found that peripheral blood mononucleated cells (PBMC) from patients or from healthy controls did not respond with cytokine production upon stimulation by Hsp60 in vitro with patterns that would differentiate patients from controls with statistical significance. This "negative” outcome appeared when the data were pooled and ana…
PGAC: A Parallel Genetic Algorithm for Data Clustering
2005
Cluster analysis is a valuable tool for exploratory pattern analysis, especially when very little a priori knowledge about the data is available. Distributed systems, based on high speed intranet connections, provide new tools in order to design new and faster clustering algorithms. Here, a parallel genetic algorithm for clustering called PGAC is described. The used strategy of parallelization is the island model paradigm where different populations of chromosomes (called demes) evolve locally to each processor and from time to time some individuals are moved from one deme to another. Experiments have been performed for testing the benefits of the parallelisation paradigm in terms of comput…
Typology of pentad circulation anomalies over the Eastern Africa - Western Indian Ocean region, and their relationship with rainfall
2005
International audience; The aim of this study was to classify the most frequently observed atmospheric circula- tion anomaly patterns in eastern Africa and the adjacent Indian Ocean. As an example of the useful- ness of such a classification, the second objective was to test whether these patterns account for intraseasonal rainfall anomalies in the region. A partitioning algorithm, known as dynamical cluster analysis, was therefore applied to the zonal (U)and meridional (V)components of the wind anom- alies, obtained from the NCEP-NCAR Reanalysis R-2 at the pentad (5 d) timescale. The 3 geopoten- tial levels 850, 700 and 200 hPa were combined. Focus is on the transition seasons (March to Ma…
Predicting lorawan behavior. How machine learning can help
2020
Large scale deployments of Internet of Things (IoT) networks are becoming reality. From a technology perspective, a lot of information related to device parameters, channel states, network and application data are stored in databases and can be used for an extensive analysis to improve the functionality of IoT systems in terms of network performance and user services. LoRaWAN (Long Range Wide Area Network) is one of the emerging IoT technologies, with a simple protocol based on LoRa modulation. In this work, we discuss how machine learning approaches can be used to improve network performance (and if and how they can help). To this aim, we describe a methodology to process LoRaWAN packets a…
Exploratory approach for network behavior clustering in LoRaWAN
2021
AbstractThe interest in the Internet of Things (IoT) is increasing both as for research and market perspectives. Worldwide, we are witnessing the deployment of several IoT networks for different applications, spanning from home automation to smart cities. The majority of these IoT deployments were quickly set up with the aim of providing connectivity without deeply engineering the infrastructure to optimize the network efficiency and scalability. The interest is now moving towards the analysis of the behavior of such systems in order to characterize and improve their functionality. In these IoT systems, many data related to device and human interactions are stored in databases, as well as I…
Gamma Knife treatment planning: MR brain tumor segmentation and volume measurement based on unsupervised Fuzzy C-Means clustering
2015
Nowadays, radiation treatment is beginning to intensively use MRI thanks to its greater ability to discriminate healthy and diseased soft-tissues. Leksell Gamma Knife® is a radio-surgical device, used to treat different brain lesions, which are often inaccessible for conventional surgery, such as benign or malignant tumors. Currently, the target to be treated with radiation therapy is contoured with slice-by-slice manual segmentation on MR datasets. This approach makes the segmentation procedure time consuming and operator-dependent. The repeatability of the tumor boundary delineation may be ensured only by using automatic or semiautomatic methods, supporting clinicians in the treatment pla…