Computationally intensive segmentation algorithms often operate on an image pre-segmented into small regions referred to as “superpixels”. We investigate the effect of the choice of the pre-segmentation algorithm and its parameters on the outcome of the final segmentation. Three pre-segmentation algorithms are compared. To avoid the particularities of sophisticated segmentation algorithms, the ...

Verbs can only be used with a few specific arrangements of their arguments (syntactic frames). Most theorists note that verbs can be organized into a hierarchy of verb classes based on the frames they admit. Here we show that such a hierarchy is objectively well-supported by the patterns of verbs and frames in English, since a systematic hierarchical clustering algorithm converges on the same s...

This work focuses on clustering a site into groups of documents that are predictive of future user accesses. Two approaches have been developed and tested. The first approach uses semantic information inherent in the documents to facilitate the clustering process. User access history is then used to reorganize the clusters iteratively so as to better indicate access patterns. This method was fo...

We present experiments aiming at an automatic classification of Spanish verbs into lexical semantic classes. We apply well-known techniques that have been developed for the English language to Spanish, proving that empirical methods can be re-used through languages without substantial changes in the methodology. Our results on subcategorisation acquisition compare favourably to the state of the...

In this paper, we propose a scheme for annotating utterance-level units in Japanese dialogs, which emerged from an analysis of the interrelationship among four schemes, i) inter-pausal units, ii) intonation units, iii) clause units, and iv) pragmatic units. The associations among the labels of these four units were illustrated by multiple correspondence analysis and hierarchical cluster analysi...

In this paper, we describe our approach and its results for MediaEval 2014 Retrieving Diverse Social Images Task. The basic idea of our proposed method is to filter out non-relevant images at the beginning of the process and then construct a hierarchical tree which allows to cluster the images with different criteria on visual and textual features. Experimental results shown that it is stable a...

We present techniques to semi-automatically discover Recurrent Visual Semantics (RVS) -the repetitive appearance of visually similar elements such as objects and scenesin consumer photographs. First, we introduce the detection of "bracketing" (very similar photographs) using an edge-correlation metric, which outperforms color histogram. Then, we use color and novel composition features (based o...

This paper focuses on the stability-based approach for estimating the number of clusters K in microarray data. The cluster stability approach amounts to performing clustering successively over random subsets of the available data and evaluating an index which expresses the similarity of the successive partitions obtained. We present a method for automatically estimating K by starting from the d...

In our previous study, task segmentation was done by mnSOM, using prior information that winner modules corresponding to subsequences in the same class share the same label. Since this prior informatiom is not available in real situation, segmentation thus obtained should be regarded as the upper bound for the performance, not as a candidate for performance comparison. Present paper proposes to...

This paper considers networks where relationships between nodes are represented by directed dissimilarities. The goal is to study methods that, based on the dissimilarity structure, output hierarchical clusters, i.e., a family of nested partitions indexed by a connectivity parameter. Our construction of hierarchical clustering methods is built around the concept of admissible methods, which are...