You are here

Home

Demo Sessions

Session Co-Chairs: Francesco Bonchi and Matthijs van Leeuwen
Demos will be held in the foyer area on the second floor.

Tuesday Morning

  1. Mining Trajectories for Spatio-temporal Analytics by Songhua Xing, Xuan Liu, Qing He and Arun Hampapur

    Mining high resolution trajectories from moving vehicles can provide insightful analytics and enable location based decision making. In this paper, we introduce a trajectory mining prototype system to generate the trajectory heat map at aggregated level for the online spatial-temporal analytics. The proposed method is scalable and efficient since we develop a mechanism using a small subset of trajectory points to capture all trajectory patterns. We experimentally verified the applicability and scalability of this system with large scale real world dataset.

  2. Mining Temporal Profiles of Mobile Applications for Usage Prediction by Zhung-Xun Liao, Tsu-Jou Shen and Wen-Chih Peng

    Due to the proliferation of mobile applications (abbreviated as Apps) on smart phones, users can install many Apps to facilitate their life. Usually, users browse their Apps by swiping touch screen on smart phones, and are likely to spend much time on browsing Apps. In this paper, we design an AppNow widget that is able to predict users' Apps usage. Therefore, users could simply execute Apps from the widget. The main theme of this paper is to construct the temporal profiles which identify the relation between Apps and their usage times. In light of the temporal profiles of Apps, the AppNow widget predicts a list of Apps which are most likely to be used at the current time. AppNow consists of three components, the usage logger, the temporal profile constructor and the Apps predictor. First, the usage logger records every App start time. Then, the temporal profiles are built by applying Discrete Fourier Transform and exploring usage periods and specific times. Finally, the system calculates the usage probability at current time for each App and shows a list of Apps with highest probability. In our experiments, we collected real usage traces to show that the accuracy of AppNow could reach 86% for identifying temporal profiles and 90% for predicting App usage.

  3. Rapid Damage eXplorer (RDX): A Probabilistic Framework for Learning Changes From Bitemporal Images by Ranga Raju Vatsavai

    Recent decade has witnessed major changes on the Earth, for example, deforestation, varying cropping and human settlement patterns, and crippling damages due to disasters. Accurate damage assessment caused by major natural and anthropogenic disasters is becoming critical due to increases in human and economic loss. This increase in loss of life and severe damages can be attributed to the growing population, as well as human migration to the disaster prone regions of the world. Rapid assessment of these changes and dissemination of accurate information is critical for creating an effective emergency response. Change detection using high-resolution satellite images is a primary tool in assessing damages, monitoring biomass and critical infrastructures, and identifying new settlements. In this demo, we present a novel supervised probabilistic framework for identifying changes using very high-resolution multispectral, and bitemporal remote sensing images. Our demo shows that the rapid damage explorer (RDX) system is resilient to registration errors and differing sensor characteristics.

Wednesday Morning

  1. Cubix: A Visual Analytics Tool for Conceptual and Semantic Data by Cassio Melo, Alexander Mikheev, Benedicte Le Grand and Marie-Aude Aufaure

    This paper presents Cubix, a Formal Concept Analysis (FCA)-based analytics tool for Business Intelligence. The main purpose of Cubix is to provide novel ways of applying visual analytics in which meaningful diagrammatic representations will be used for manipulating, filtering and visually querying complex data. We present its main features, typical applications and future steps towards an advanced FCA-based visual analytics.

  2. A Subspace Clustering Extension for the KNIME Data Mining Framework by Stephan Günnemann, Hardy Kremer, Richard Musiol, Roman Haag and Thomas Seidl

    Analyzing databases with many attributes per object is a recent challenge. For these high dimensional data it is known that traditional clustering algorithms fail to detect meaningful patterns. As a solution subspace clustering techniques were introduced. They analyze arbitrary subspace projections of the data to detect clustering structures. In this demonstration, we introduce the first subspace clustering extension for the well-established KNIME data mining framework. While KNIME offers a variety of data mining functionalities, subspace clustering is missing so far. Our novel extension provides a multitude of algorithms, data generators, evaluation measures, and visualization techniques specifically designed for subspace clustering. It deeply integrates into the KNIME framework allowing a flexible combination of the existing KNIME features with the novel subspace components. The extension is available on our website.

  3. Supporting the Discovery of Relevant Topological Patterns in Attributed Graphs by Julien Salotti, Marc Plantevit, Celine Robardet and Jean-Francois Boulicaut

    We propose TopGraphVisualizer, a tool to support the discovery of relevant topological patterns in attributed graphs. It relies on a new pattern detection method that crucially needs for sophisticated post processing and visualization. A topological pattern is defined as a set of vertex attributes and topological properties (i.e., properties that characterize the role of a vertex within a graph) that strongly co-vary over the vertices of the graph. For instance, such a pattern in a co-authorship attributed graph where vertices represent authors, edges encode coauthor ship, and vertex attributes reveal the number of publications in several journals, could be "€œthe higher the number of publications in IEEE ICDM, the higher the closeness centrality of the vertex within the graph. Two different ways of navigation through the topological patterns and the related graph data are provided to the end-user. We exploit graph visualization and exploration techniques from the open platform Gephi. As an illustrative scenario, we consider a co-autorship attributed graph built from DBLP digital library and a video has been produced that describe the main possibilities of the TopGraphVisualizer software.

Thursday Morning

  1. Topick: Accurate Topic Distillation for User Streams by Anton Dimitrov, Alexandra Olteanu, Luke McDowell and Karl Aberer

    Users of today's information networks need to digest large amounts of data. Therefore, tools that ease the task of filtering the relevant content are becoming necessary. One way to achieve this is to identify the users who generate content in a certain topic of interest. However, due to the diversity and ambiguity of the shared information, assigning users to topics in an automatic fashion is challenging. In this demo, we present Topick, a system that leverages state of the art techniques and tools to automatically distill high-level topics for a given user. Topick exploits both the user stream and her profile information to accurately identify the most relevant topics. The results are synthesised as a set of stars associated to each topic, designed to give an intuition about the topics encompassed in the user streams and the confidence in the results. Our prototype achieves a precision of 70% or more, with a recall of 60%, relative to manual labeling. Topick is available at http://topick.alexandra.olteanu.eu.

  2. OpinioNetIt: A Structured and Faceted Knowledge-base of Opinions by Rawia Awadallah, Maya Ramanath and Gerhard Weikum

    We propose a demonstration of our system, OpinioNetIt, a structured, faceted, knowledge-base of opinions, and its use in political analysis. OpinioNetIt consists of information about people, topics and opinions in the form of triples, indicating the opinion of a person on a topic. Our focus is on acquiring opinions held by various stakeholders on politically controversial topics. Our system can be used for various kinds of analyses such as generating heat maps showing political bias, identifying flip-flopping politicians, and identifying dissenters, etc. In this demonstration proposal, we give a brief overview of our system and the proposed demonstration.

  3. AD-MAD: Integrated System for Automated Development and Optimization of Online Advertising Campaigns by Stamatina Thomaidou, Konstantinos Leymonis, Kyriakos Liakopoulos and Michalis Vazirgiannis

    Creating and monitoring a competitive and cost-effective pay-per-click advertisement campaign through the web-search channel is a resource demanding task in terms of human expertise and effort. Assisting or even automating the work of an advertising specialist will have an unrivaled commercial value. In this demonstration we present a prototype and a functional web application for semi- and fully-automated creation, monitoring, and management of cost-efficient pay-per-click campaigns with budget constraints. The prototype is experimentally evaluated on real world Google Ad Words campaigns and shows a promising behavior with regards to campaign performance statistics outperforming systematically the competitive manually created and/or monitored campaigns.