Corpus ID: 32477447

Apprentissage non supervisé de séries temporelles à l'aide des k-means et d'une nouvelle méthode d'agrégation de séries

  title={Apprentissage non supervis{\'e} de s{\'e}ries temporelles {\`a} l'aide des k-means et d'une nouvelle m{\'e}thode d'agr{\'e}gation de s{\'e}ries},
  author={R. Gaudin and N. Nicoloyannis},
A transformer for measuring current flowing through a conductor includes a U-shaped housing which is adapted to be received about the conductor. A closure member is secured by a sliding hinge assembly to one distal end of the housing, and is adapted to be latched between the distal ends. Within the housing a plurality a U-shaped core laminations are disposed in adjacent laminated groups. The closure member includes a plurality of laminated groups which are disposed to interleave with the distal… Expand
Reconnaissance automatique des grapho-éléments temporels de l'électroencéphalogramme du sommeil
En neurophysiologie, l'electroencephalogramme (EEG) constitue un moyen d'etude important dans divers domaines parmi lesquels nous pouvons citer le diagnostic des troubles du sommeil et l'analyse deExpand
Paving the way for next generation data-stream clustering: towards a unique and statistically valid cluster structure at any time step
This work focuses here on appending a Monte-Carlo method for extracting statistically valid inter-text links, which looks promising applied both to an excerpt of the Pascal bibliographic database, and to the Reuters-RCV1 news test collection. Expand
Clustering of Bi-Dimensional and Heterogeneous Time Series: Application to Social Sciences Data
The Longest Common Subsequence time series distance is used for its efficiency to manage time stretching and it is extended to the bidimensional and heterogeneous case and gives some pertinent and surprising clusters that can be easily analyzed by sociologists. Expand
How to use ants for data stream clustering
This paper presents a new bio-inspired algorithm that dynamically creates groups of data based on the concept of artificial ants that move together in a complex manner with simple localization rules, and suggests an extension to this algorithm to treat data streaming. Expand
Incremental clustering of data stream using real ants behavior
A new biomimetic method nammed CL-AntInc for data incremental clustering that uses the behavior of real ants and deals with the issue of data volume through a clustering heuristic. Expand


Derivative Dynamic Time Warping
Dynamic time warping (DTW), is a technique for efficiently achieving this warping of sequences that have the approximately the same overall component shapes, but these shapes do not line up in X-axis. Expand
Fast Similarity Search in the Presence of Noise, Scaling, and Translation in Time-Series Databases
We introduce a new model of similarity of time sequences that captures the intuitive notion that two sequences should be considered similar if they have enough non-overlapping time-ordered pairs ofExpand
Discovering similar multidimensional trajectories
This work formalizes non-metric similarity functions based on the longest common subsequence (LCSS), which are very robust to noise and furthermore provide an intuitive notion of similarity between trajectories by giving more weight to similar portions of the sequences. Expand
An Enhanced Representation of Time Series Which Allows Fast and Accurate Classification, Clustering and Relevance Feedback
We introduce an extended representation of time series that allows fast, accurate classification and clustering in addition to the ability to explore time series data in a relevance feedbackExpand
UCR Time Series Data Mining Archive
A silicon carbide body with a density of at least 85% is machined to required specification and then fired at a temperature ranging from 1400 DEG C to 2100 DEG C in a firing atmosphere ranging inExpand
Indexing multi-dimensional time-series with support for multiple distance measures
The experimental results demonstrate that the index motivated by the need for a single index structure that can support multiple distance measures can help speed-up the computation of expensive similarity measures such as the LCSS and the DTW. Expand
Iterative Incremental Clustering of Time Series
A novel anytime version of partitional clustering algorithm, such as k-Means and EM, for time series, that works by leveraging off the multi-resolution property of wavelets and is much faster than its batch counterpart. Expand
Making Time-Series Classification More Accurate Using Learned Constraints
This work targets the accuracy aspect of DTW performance and introduces a new framework that learns arbitrary constraints on the warping path of the DTW calculation and speeds up DTW by a wide margin. Expand
Matching and indexing sequences of different lengths
This paper proposes an indexing scheme which is totally based on lengths and relative distances between sequences, and uses vp-trees as the underlying distance-based index structures in its method. Expand
Finding Motifs in Time Series
An efficient motif discovery algorithm for time series would be useful as a tool for summarizing and visualizing massive time series databases and could be used as a subroutine in various other data mining tasks, including the discovery of association rules, clustering and classification. Expand