News & Events
Danai in the news!
Danai and the Microsoft researchers Paul Bennett and Eric Horvitz studied information-seeking behavior and access to alternative versus reinforcing viewpoints for strongly polarizing topics. In particular, they analyzed the search and browsing behavior on the gun control debate following a shocking news event, the Sandy Hook Elementary School shootings in Newtown, Connecticut. The paper is at: http://arxiv.org/pdf/1405.1486v1.pdf The MIT Tech Review article is at: http://www.technologyreview.com/view/527311/sandy-hook-the-gun-control-debate-and-the-insidious-influence-of-the-filter-bubble/ Other press coverage includes: http://www.technology.org/2014/05/13/ideological-internet-bubbles-stay-strong-study-shows/# Read More
Danai invited to Heidelberg Laureate Forum
The Forum brings together winners of the Turing Award and Nevanlinna Prize (CS), as well as the Abel Prize and Fields Medal (math) for a week in Heidelberg. An international committee of experts selected 100 successful CS candidates from more than 40 countries around the world. Congratulations, Danai! Read More
DB Seminar: Atreyee Maiti
Transactions in on-line transaction processing (OLTP) workloads typically have the following characteristics: (1) they are short-lived, (2) they work on a small subset of the data, (3) they are repetitive. Traditional disk-based database management systems (DBMS) incur too much overhead for OLTP datasets that could simply be memory resident. This is because of the presence of heavyweight concurrency control and recovery mechanisms. Given the OLTP application characteristics, poor performance of traditional disk-based systems and the large amount of main memory Read More
DB Seminar: Jay-Yoon Lee
How can we visualize billion-scale graphs? How to spot outliers in such graphs quickly? Visualizing graphs is the most direct way of understand- ing them; however, billion-scale graphs are very difficult to visualize since the amount of information overflows the resolution of a typical screen. In this paper we propose NET-RAY, an open-source package for visualization- based mining on billion-scale graphs. NET-RAY visualizes graphs using the spy plot (adjacency matrix patterns), distribution plot, and correlation plot which in- volve careful Read More
Miguel, Stephan, Vagelis, Danai: Win ‘Best Student Paper’ Distinction in PAKDD’14.
SCS team and BBN and ARL collaborators win 'best student paper runner up' award, at PAKDD'14. The citation is: [bibtex file=citations.bib key=Araujo14 remove-tag=project] The paper describes a novel algorithm to spot time-evolving communities in large datasets. PAKDD is one of the top data mining conferences. The award ceremony will be on May 15, in Tainan, Taiwan. Congratulations! Read More
Mike Cafarella (University of Michigan)
Trained systems that apply machine learning to very large datasets, such as web search and IBM's Watson question-answering system, are among the most important and sophisticated software systems being constructed today. Such trained systems are frequently based on supervised learning tasks that require features, signals extracted from the data that distill complicated raw data objects into a small number of salient values. For example, a good feature for a search engine's relevance ranker might be the number of times the Read More
Database Group Meeting (April 21, 2014)
How can we correlate the neural activity in the human brain as it responds to typed words, with properties of these terms (like ’edible’, ’fits in hand’)? In short, we want to find latent variables, that jointly explain both the brain activity, as well as the behavioral responses. This is one of many settings of the Coupled Matrix- Tensor Factorization (CMTF) problem. Can we accelerate any CMTF solver, so that it runs within a few minutes instead of tens of Read More
DB alumni mentioned in the MIT technology review!
Prof. Hanghang TONG (MLD phd) and Prof. Leman Akoglu (CSD phd) worked on analyzing the correlation between quality of questions and quality of answers, in "stack-overflow" - their paper is on the web: http://arxiv.org/abs/1311.6876 The MIT-tech-review article can be found at http://www.technologyreview.com/view/522171/data-mining-reveals-the-secret-to-getting-good-answers/ Congratulations, Hanghang and Leman! Read More
U Kang is chosen as finalist, for the KDD-dissertation award!
Prof. U Kang, SCS Alumni, is chosen as the finalist for the 2013 KDD Dissertation Award, which is the highest honor for a data mining thesis. His dissertation is titled "Mining Tera-Scale Graphs: Theory Read More
SCS team, alumni and colleagues attract the Best Paper award in CIKM’12!
The paper gives fast, scalable algorithms to strengthen (or weaken) a network, by careful edge additions (or removals), and gives applications on real graphs. The full citation is: [bibtex file=citations.bib key=Tong2012a remove-tag=project] The first two authors are SCS alumni (2010 and 2012, resp.), advised by the last author, SCS faculty Prof. Christos Faloutsos. CIKM is one of the top conferences in information retrieval and data mining with 1099 articles submitted this year, and only 146 of them accepted as full Read More