News
ROBUST Technology Helps to Experience the Volvo Ocean Race
Together with other researchers, the ROBUST team at NUIG started to work on a demo application for the Volvo Ocean Race, which will finish it's final leg in June in Galway, Ireland. The application will leverage ROBUST technology to combine real-time network analysis on social media streams with a neat handheld user interface to improve visitor experience. It will showcase the market potential of ROBUST technology and act as a basis for further developments of advanced analytical services. Stay tuned and plan your trip to Galway!
Presentation of ROBUST at WWW 2012 in Lyon
The ROBUST consortium has sponsored the 21st World Wide Web conference in Lyon and used the occasion to present the project, its setting, latest results and show a first demo of the ROBUST platform. The World Wide Web conference is the major event in the field of research and developments in the context of the Web. It took place in Lyon from April 16th until April 20th and attracted nearly 2,000 visitors.
TEMIS workshop on "Rule-based, Thesaurus-based or Machine Learning-based: Difference Approaches to Semantic Content Enrichment"
On May 8-10 2012, ROBUST partner TEMIS will participate at the Informare! conference in Berlin with a talk and a workshop from Stefan Geißler: "Rule-based, hesaurus-based or Machine Learning-based: Difference Approaches to Semantic Content Enrichment". The talk outlines analysis options of the processing engine which is part of the ROBUST VM and which is used to perform some of the text-based analysis tasks in the project. Find more information at the conference website.
Harnessing Explicit Semantic Analysis to Detect Taxonomy Prevalence
In an attempt to improve its taxonomy overlays, Polecat has been investigating Explicit Semantic Analysis [1,2] – a suggestion from the University of Koblenz Landau. The original idea behind ESA is to allow a user to measure the semantic similarity between two documents.
Publications on Cross Community Influence and Information Diffusion
Václav Belák, Samantha Lam and Conor Hayes from NUIG got two ROBUST-related publications accepted: 'Cross-Community Influence in Discussion Fora', to appear at ICWSM'12, proposes an extensible framework for scalable analysis of cross-community influence in discussion fora.
Why Your Company Should be Interested in Social CRM
ROBUST partner Software Mind offers Customer Relationship Management products based on Social CRM strategy. Here are a couple of facts describing companies’ growing interest in a brand new channel of communication with customers, that is social media.
Towards a Metaphor Visualization
Over the last few months, ROBUST partner Polecat has continued to look into the concept of a metaphor based visualisation. This is more ambitious in scope than using metaphors within the visualisation; instead it is an attempt to create a visualisation that exclusively uses a single metaphor.
Using KONECT to Investigate Graph Sampling Techniques
The University of Koblenz-Landau is currently investigating graph sampling techniques for very large networks. The networks studied in the ROBUST projects such as the business communities of IBM and SAP have such a large size that most conventional network mining algorithms cannot be applied to them easily: One must use clusters of servers to process the whole networks, necessitating huge amounts of computing power, memory and network bandwidth.
Active Dissemination of ROBUST Results at NUIG
Researchers from DERI, NUI Galway submitted three papers on ROBUST topics to the 6TH International AAAI Conference on Weblogs and Social Media (ICWSM 2012). The topics range from behavioural clustering of users for dynamic role analysis, over thread reconstruction for online discussion forums from different domains, to cross-community influence in online forums. Another work on maximising information spread in online forums will be submitted to the Mining Social Network Dynamics workshop at the World Wide Web conference 2012.
Diversity among Enterprise Online Communities
There is a growing body of research into the adoption and use of social software in enterprises. However, less is known about how groups, such as communities, use and appropriate these technologies, and the implications for community structures. In a study of 188 very active online enterprise communities in the internal IBM Connections deployment in IBM, systematic differences in size, demographics and participation, aligned with differences in community types were found.
CORMSIS Develops New Graph Metrics
Online communities are invariably expressed as graphs, ie. as collections of nodes with connections (edges) between each other. An important measure on such a graph is the distance between two nodes, which is often expressed in terms of a shortest route between the given nodes.
Erik Aumayr joins ROBUST Team of NUIG
Erik Aumayr submitted his Diploma thesis on "Reconstruction of Threaded Conversations in Online Discussion Forums". This work was done in the context of the ROBUST project in a collaboration between DERI, NUI Galway and the Ilmenau University of Technology. Erik will soon start as a PhD student in DERI and will focus on similar ROBUST topics.
ROBUST contributes to Apache Giraph
A key focus of ROBUST is the analysis of huge social graphs. For general-purpose big data computation, the map-reduce computing model has been well adopted and the most deployed map-reduce infrastructure is Apache Hadoop. However the map-reduce paradigm is not well suited for processing such large graphs. One of the core tasks of research in the ROBUST project is to identify more suitable and powerful paradigms for implementing large scale graph algorithms.
Trilateral meeting in Milton Keynes
Members from the ROBUST project from IBM, OU, and NUIG got together in Milton Keynes for a two day meeting on 11th and 12th of January to discuss progress and plan joint work. The meeting was focused around the requirements of the IBM use case. Main topics discussed were (a) impact of community type on its activity patterns and behaviour, (b) semantics and models for behaviour analysis, and (c) advanced techniques and algorithms for churn analysis. The meeting was highly effective and resulted in specific joint work and dissemination plans for the next 6 months of ROBUST.
Talk on "The Intelligence of Social Connections... and why this is relevant for P2P"
On 14.12.2011, Marcel Karnstedt gave an invited talk on ROBUST topics in the context of a lecture on P2P Networks at TU Darmstadt, Germany. The title of the talk was "The Intelligence of Social Connections ... and why this is relevant for P2P".
Alice Hogan has joined the DERI team at NUIG to work on the ROBUST project as a research assistant
Alice has a BSc in Experimental Physics from NUIG, a Higher Diploma in Modelling and Numerical Computing from UCC and a Masters in Electronic Engineering from NUI Maynooth. Alice has also worked in various roles in industry including Defect Analysis Engineer in Intel, Kildare, Quality Engineer in Boston Scientific, Galway, Quality Engineer in RR Donnelly, Limerick and most recently worked as a Junior Developer/Process Engineer in Hewlett-Packard, Galway.
WeST Team Becomes Semantic Web World Champion
The SchemEX system presented by Mathias Konrath, Dr. Thomas Gottron and Jun.-Prof. Dr. Ansgar Scherp from the Institute WeST has won the Semantic Web Challenge in the Billion Triples category at the 10th International Semantic Web Conference, ISWC 2011. SchemEX provides a smart approach for indexing large and distributing graph based data as they appear also in the context of online communities. With Thomas Gottron, also researchers from the ROBUST project have contributed to this success.
ROBUST initiates the Virtual Center of Excellence on Online Communities
Online communities are in the focus of several current research projects. While the focus and aims are always different, some key questions might be of interest across several of these projects. In order to foster the exchange of findings and to discuss ideas ROBUST initiated the Virtual Center of Excellence on Online Communities.
TUB aims to process a 700 million interactions dataset for ROBUST
TU Berlin is currently continuing its work on distributed link prediction and collaborative filtering algorithms. At the heart of these approaches is a pairwise similarity comparison of the columns of an enormous sparse matrix.
Overview of Visualizing Techniques for Online Communities
As part of the next stage of the robust deliverables, Polecat is starting to look at creating a visual metaphor to represent and monitor community health. This still appears to be an underdeveloped area of research, and there are few really pure examples around. Most data still tends to be represented with traditional charts and graphs.
Forgot your password?
New user?
