Visualizing Research Works in the Water Resources Industry

With the increasing practice of making data openly available, nowadays there is a growing amount of information easily available pertaining to water resources and ecology. Scientific works by researchers across the world contribute to the abundance in such data. Major challenges that emerge due to the volume of data include the discovery of useful and relevant content, as well as learning and interpretation of the various disparate content. In this work, we aim to aid researchers and interested stakeholders in understanding the vast landscape of scientific research in the water resources industry. We integrate different sources of data from the Web; journals from Elsevier, tweets from Twitter, and wikipedia annotations. We use interactive visualizations in order to engage the users and satisfy their information needs.

Online prototype: http://www.l3s.de/~kawase/vici/

Exploiting the Wisdom of the Crowds for Characterizing and Connecting Heterogeneous Resources

Ricardo Kawase presenting @HT2014 in Santiago, Chile (picture taken by Christoph Trattner)

Heterogeneous content is an inherent problem for cross-system search, recommendation and personalization. In this paper we investigate differences in topic coverage and the impact of topics in different kinds of Web services. We use entity extraction and categorization to create fingerprints that allow for meaningful comparison. As a basis taxonomy, we use the 23 main categories of Wikipedia Category Graph, which has been assembled over the years by the wisdom of the crowds. Following a proof of concept of our approach, we analyze differences in topic coverage and topic impact. The results show many differences between Web services like Twitter, Flickr and Delicious, which reflect users’ behavior and the usage of each system. The paper concludes with a user study that demonstrates the benefits of fingerprints over traditional textual methods for recommendations of heterogeneous resources.

Authors: Ricardo Kawase, Patrick Siehndel, Bernardo Pereira Nunes, Eelco Herder and Wolfgang Nejdl

PDF: kawase-ht2014

Online Prototype: http://twikime.l3s.uni-hannover.de

 

DBLPXplorer: Interactive Graphical Interfaces for the Computer Science Bibliography – SHORTLISTED – PEOPLE’S CHOICE WINNER

Winners of the LinedUp Vidi challenge @ESWC2014

 

Vidi, the second of three consecutive competitions, sponsored and organized by the LinkedUp Project, had the goal to gather interesting and innovative tools and applications that analyze and/or integrate open web data for educational purposes.

With this goal in mind, L3S team of researchers (Ricardo Kawase, Ujival Gadiraju and Patrick Siehndel) proposed the DBLPXplorer, a set of interactive visualization tools to facilitate the browsing and discovery of scientific work. The work was awarded the People’s Choice Winner – a category where the winner was decided by the audience voting.

DBLPXplorer: http://linkedup-project.eu/2014/04/16/dblpxplorer/

Vidi Winners Announced: http://linkedup-project.eu/2014/05/29/vidi-winners-announced/

LinkedUp Project: http://linkedup-project.eu/

DBLPXplorer: Interactive Graphical Interfaces for the Computer Science Bibliography

Ricardo Kawase presenting @ESWC2014 LinkedUp – Vidi Challenge

Every year thousands of new research works are indexed and published online. Scientific publications involve mainly two sets of actors; namely, authors and articles. Consequently, a huge tangle of relations emerge together, where authors collaborate with several other authors and articles reference past literature. Due to this complex network, keeping up to date with the latest research in a particular field is often a time consuming task. Currently, available tools to explore such information are solely text based. The information seeker has to search, browse and navigate page by page in order to find relevant research. Yet, one cannot harness an overview of underlying networks and connections. At the same time, there is an abundance of information in the form of nearly disjoint datasets relevant to research and the actors involved in the Linked Open Data cloud. To facilitate the exploration of authors, scientific research and relations, we propose a visual exploratory interface for DBLP Computer Science Bibliography. To further enrich the data we extract authors’ keywords from the articles and additionally annotate each article with identified DBPedia entities. The presentation layer consists of several user friendly exploratory interfaces that utilize state of the art javascript library D3 (Data-Driven Documents). Our interfaces include overview of particular venues, authors’ profiles, scientific articles, relations and a knowledge base of keywords and semantic annotations. To complete our work, we expose all the enriched data as linked data. – See more at: http://linkedup-challenge.org/vidi/#DBLPXplorer

Authors: Ricardo Kawase, Ujwal Gadiraju and Patrick Siehndel

Demo: http://www.l3s.de/~kawase/DBLPXplorer/

Haters gonna hate: job-related offenses in Twitter

In this paper, we aim at finding out which users are likely to publicly demonstrate frustration towards their jobs on the microblogging platform Twitter – we will call these users haters. We show that the profiles of haters have specific characteristics in terms of vocabulary and connections. The implications of these findings may be used for the development of an early alert system that can help users to think twice before they post potentially self-harming content.

Authors: Ricardo Kawase, Patrick Siehndel, and Eelco Herder

PDF: kawase-www2014

FireMe! Website: http://fireme.l3s.uni-hannover.de/

Who wants to get fired?

Ricardo Kawase presenting @WEBSCI2013

Microblogging services like Twitter have witnessed a flood of users and short updates (tweets). Although this phenomenon brings new possibilities of communication, it also brings dangerous consequences. From time to time, people post tweets guided by strong emotions. By default, tweets are public and anyone, anywhere can instantly see your updates, creating high exposure and lack of awareness about privacy issues. In many cases, this may lead to consequences that can be harmful to one’s personal and professional life. In this paper, we investigate the posting behavior of people who tweet that they hate their jobs and bosses and their responses to alerts about the potential damage that such a tweet may cause. We show that, in many cases, people are not aware about the dimension of their audience, and once alerted, they often regret what they have publicly said. Our analysis leads us to believe that many users could benefit from a ‘give a second thought before posting’ tool that may save their jobs.

Authors:  Ricardo Kawase, Bernardo Pereira Nunes, Eelco Herder, Wolfgang Nejdl and Marco Antonio Casanova

PDF: kawase-websci2013

FireMe! Website: http://fireme.l3s.uni-hannover.de/

TwikiMe! User profiles that make sense.

The use of social media has been rapidly increasing in the last years. Social media, such as Twitter, has become an important source of information for a variety of people. The public availability of data describing some of these social networks has led to a great deal of research in this area. Link prediction, user classification and community detection are some of the main research areas related to social networks. In this paper, we present a user modeling framework that uses Wikipedia to model user interests inside a social network. Our model of user interests reflects the areas a user is interested in, as well as the level of expertise a user has in a certain field.

Authors: Patrick Siehndel, Ricardo Kawase

PDF: siehndel-iswc2012

Online Prototype: http://twikime.l3s.uni-hannover.de/twikime.php