This paper presents a new bio-inspired algorithm (FClust) that dynamically creates and visualizes groups of data. This algorithm uses the concepts of a flock of agents that move together in a complex manner with simple local rules. Each agent represents one data. The agents move together in a 2D environment with the aim of creating homogeneous groups of data. These groups are visualized in real time, and help the domain expert to understand the underlying structure of the data set, like for example a realistic number of classes, clusters of similar data, isolated data. We also present several extensions of this algorithm, which reduce its computational cost, and make use of a 3D display. This algorithm is then tested on artificial and real-world data, and a heuristic algorithm is used to evaluate the relevance of the obtained partitioning.
3D drawing problems of the 90s were essentially restricted on representations in 3D perspective. However, recent technologies offer 3D stereoscopic representations of high quality which allow the introduction of binocular disparities, which is one of the main depth perception cues, not provided by the 3D perspective. This paper explores the relevance of stereoscopy for the visual identification of communities, which is a task of great importance in the analysis of social networks. A user study conducted on 35 participants with graphs of various complexity shows that stereoscopy outperforms 3D perspective in the vast majority of the cases. When comparing stereoscopy with 2D layouts, the response time is significantly lower for 2D but the quality of the results closely depend on the graph complexity: for a large number of clusters and a high probability of cluster overlapping stereoscopy outperforms 2D whereas for simple structures 2D layouts are more efficient.
International audienceAggregating statistical representations of classes is an important task for current trends in scaling up learning and recognition, or for addressing them in distributed in- frastructures. In this perspective, we address the problem of merging probabilistic Gaus- sian mixture models in an efficient way, through the search for a suitable combination of components from mixtures to be merged. We propose a new Bayesian modelling of this combination problem, in association to a variational estimation technique, that handles efficiently the model complexity issue. A main feature of the present scheme is that it merely resorts to the parameters of the original mixture, ensuring low computational cost and possibly communication, should we operate on a distributed system. Experimental results are reported on real dat
In this paper, we propose an approach to interactive navigation in image collections. As structured groups are more appealing to users than flat image collections, we propose an image clustering algorithm, with an incremental version that handles time-varying collections. A 3D graph-based visualization technique reflects the classification state. While this classification visualization is itself interactive, we show how user feedback may assist the classification, thus enabling a user to improve it.
Axes are the main components of coordinate systems representations. They play a critical role for the visual analysis of multi-dimensional data. However their representation seems to have always be considered self evident, with oriented lines crossing at an origin, completed with labels such as ticks and names. Such classical representation show limits when it comes 3D visualization and immersive analytics (IA), mainly because orthogonal projection of points on linear axes is hard in a 3d environment, and because the user can move and the axes get out of her field of view. In this paper we propose a taskbased definition of axes and coordinate systems representation, as well as a tentative design space for coordinates systems representation in immersion. We also present an exploratory user study we carried out to compare three grid-based representations of coordinate systems for multidimensional data analysis with 3D scatterplots.
This paper proposes a solution to the problem of aggregating versatile probabilistic models, namely mixtures of probabilistic principal component analyzers. These models are a powerful generative form for capturing high-dimensional, non Gaussian, data. They simultaneously perform mixture adjustment and dimensionality reduction. We demonstrate how such models may be advantageously aggregated by accessing mixture parameters only, rather than original data. Aggregation is carried out through Bayesian estimation with a specific prior and an original variational scheme. Experimental results illustrate the effectiveness of the proposal.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.