The Data Analysis and Biomedical Analytics Center extends a warm welcome to Mladen Nikolic who joined Prof. Obradovic's lab on November 5th, 2012. Mladen Nikolic is a PhD candidate and a teaching assistant at the Faculty of Mathematics, University of Belgrade, Serbia. As a visiting scholar at Prof. Obradovic's laboratory, he is to engage in the research conducted in "Prospective Analysis of Large and Complex Partially Observed Temporal Social Networks" project.
Prof. Zoran Obradovic co-chairs 2013 SIAM International Conference on Data Mining (with Prof. J. Ghosh from U. Texas).
Prof. Zoran Obradovic co-chairs 2013 SIAM International Conference on Data Mining (with Prof. J. Ghosh from U. Texas). The conference will be held on May 2-4, 2013 at Austin, Texas. SDM2013 papers submission deadline is Oct. 12, 2012. For additional information and call for papers, see http://www.siam.org/meetings/sdm13/
The Data Analysis and Biomedical Analytics (DABI) Center welcomes Assistant Professor Gregor Stiglic who joined Professor Obradovic's lab on September 19, 2012. Dr. Stiglic will work at DABI as a visiting scholar on a DARPA funded project, developing spatio-temporal algorithms for analysis of healthcare related problems. The algorithms will be extended to analysis of large scale data from social networks. Dr. Stiglic is an assistant professor at the Faculty of Health Sciences, University of Maribor in Slovenia. His main research interests include machine learning techniques with application to bioinformatics and healthcare. Specific areas of technical interest include comprehensibility of classifiers, human interaction based classification, stability of feature selection algorithms, meta-learning and spatio-temporal rule discovery.
The Data Analysis and Biomedical Analytics Center extends a warm welcome to Jelena Slivka who joined Prof. Obradovic's lab on October 1st, 2012. Jelena Slivka is a third year PhD student and teaching assistant at the Faculty of Technical Sciences, University of Novi Sad, Serbia. As a visiting scholar at Prof. Obradovic's laboratory, she is to engage in the research conducted in "Prospective Analysis of Large and Complex Partially Observed Temporal Social Networks" project.
The Big Data TV interview with Dr. Ken Blank (senior vice provost for research and education) and Dr Zoran Obradovic (Data Analytics and Biomedical Informatics Center Director; Computer and Information Science Professor)
The Big Data TV interview with Dr. Ken Blank (senior vice provost for research and education) and Dr Zoran Obradovic (Data Analytics and Biomedical Informatics Center Director; Computer and Information Science Professor) will be broadcast on Tuesday 9/25/12 at 12:30pm and 8:30pm on TUTV. TUTV is seen on Comcast Channel 50 and Verizon Channel 45 within the City of Philadelphia. The program is also available LIVE online, anywhere, at those same times at http://www.templetv.net/watch-live/
Dr. Zoran Obradovic has been awarded another exciting DARPA grant related to big data analysis. For this four year $2,907,908 project started in August2012, entitled âProspective Analysis of Large and Complex Partially Observed Temporal Social Networksâ Prof. Obradovic is the principal investigator and co-investigators are Prof. Emily Fox at University of Washington Statistics Department and Prof. Katya Scheinberg at Lehigh University Industrial and Systems Engineering Department. The analysis of social networks often assumes a time invariant scenario, while in practice actor attributes and links in such networks evolve over time and are inextricably dependent on each other. In addition, the temporal graph is just partially observed, multiple kinds of links exist among actors, various actors have different temporal dynamics and environmental influence can be both positive and negative. This project will closely examine the hypothesis that a unified approach of jointly modeling these and related problems is beneficial for prospective analysis of large-scale partially observed temporal hypergraphs.
Starting September 2012 Prof. Zoran Obradovic will lead a GlaxoSmithKline funded project entitled "Predicting Drug Indications by Integrating Multiple Data Sources." Finding new indications for existing drugs, or drug repositioning, has been used as a strategy for decades to get drugs to more patients. To predict drug indications for both approved drugs and novel molecules, this project proposes a new approach based on the observation that similar drugs are indicated for similar diseases, and utilizes multiple drug-drug and/or disease-disease similarity measures for the prediction task. Chemical similarity has been considered as the most widely used metric for drug indications prediction, following the hypothesis that drug compounds with similar structures and chemical properties have similar use as a therapy. This study will explore to what extent the similarity from other sources (e.g., clinical side-effects, drug targets, gene expression, and bioassay) also implies similar drug indications. Furthermore, the project will study whether combining the similarity metrics from multiple sources can increase the accuracy of drug indications prediction.
Professor Yates has received a new 3 year NSF grant entitled "RI: Small: Learning Open Domain Semantic Parsers". Following is the scope of this $425,821 single investigator grant.
Supervised semantic parsers, which learn to map language to relational data, perform poorly on texts that differ in vocabulary or style from the training text, and on databases that differ from the database used in training. Today's semantic parsers have only been tested on narrowly-circumscribed domains like geography, but the ideal semantic parser would generalize to the many and incredibly diverse relational databases available on the Web.
This project develops semantic parsers that approach this ideal system. The project divides the overall task into two parts: mapping named-entities in text to database constants in any domain, and mapping full sentences and questions to logical forms written in a variant of the lambda calculus. Techniques for resolving named-entities make use of domain-independent contextual information around the named-entity for disambiguation. To connect words like "directed" with a database relation listing directors of movies, the project relies on schema-matching techniques from database integration. The system extracts a relational view of a corpus, and then generates alignments between these extracted alignments and the fixed relational structure of existing databases. The project uses transfer-learning and co-training approaches to estimate parameters for statistical models for named-entity disambiguation and schema matching across domains and databases.
Prof. Yuhong Guo and her student Suicheng Gu received Outstanding Paper Award at 26th AAAI Conference on Artificial Intelligence. Their awarded paper is entitled "Learning SVM Classifiers with Indefinite Kernels."
Dr. Obradovic's sepsis modeling project in the news
Dr. Zoran Obradovic is co-editing the Applications of Computational Intelligence in Health Informatics issue of Advances in Artificial Neural Systems.
Dr. Zoran Obradovic is co-editing the Applications of Computational Intelligence in Health Informatics issue of Advances in Artificial Neural Systems. Manuscripts are due by May 4, 2012 and the special issue will be published in September 2012. For more details see the following link.
Dr. Zoran Obradovic is giving an invited lecture at IEEE International Conference on Bioinformatics and Biomedicine 2011. His lecture
"Analysis and Integration of Inconsistent and Unreliable Biomedical Prediction Models" is based on joint papers with his PhD. students Mohamed Ghalwash (in press at Molecular Biosystems journal) and Ping Zhang (ECML PKDD 2011).
Following is the abstract of this study.
In biomedical applications, multiple predictors are often developed for the same problem using multiple training datasets of various qualities. Selecting a single model from such a collection based on accuracy evaluation on a small biased set of annotated data is not necessarily the best strategy when the objective is large scale application of the model. In this talk we will discuss how to address this problem by uncertainty analysis in the reference models and in data. In addition, we will present an iterative algorithm for integrating predictions of multiple models without relying on any annotated data. The proposed solutions will be illustrated on the problem of predicting intrinsic disorder in proteins that lack a stable tertiary structure but still have important biological functions.
October, 5. 2011.
The Data Analysis and Biomedical Analytics (DABI) Center extends a warm welcome to Prof. Boris Delibasic who joined Prof. Obradovic's lab on October 1, 2011. Prof. Delibasic received a prestigious Fulbright fellowship to work as a visiting scholar at Prof. Obradovic's laboratory at DABI until June 2012. Their research objectives are to design spatio-temporal algorithms for analysis of ski injuries and to discover ski injury patterns that could be used for injury prevention. They will extend the algorithms developed for ski injuries to analyze large scale data on road traffic accidents.
Dr. Delibasic is an assistant professor at the University of Belgrade in Serbia. His main research interests are data mining, decision support systems, business intelligence, and decision theory. Dr. Delibasic is also an adjunct lecturer at the University of Jena in Germany. With Prof. Obradovic he has already published several research articles on design of white-box data mining algorithms.
Dr Rolf Lakaemper released a Kinect TCP related website aimed at saving the burden of programming in .NET
KinectTCP is a TCP/IP server that offers all video, depth and skeleton services of Microsoft's Win7 Kinect SDK, independent from specific programming languages. kinectTCP allows programmers, who do not want to deal with windows specifics of libraries and languages, to utilize the Kinect using their language of choice, e.g. JAVA, BASIC, MATLAB, PYTHON, FLASH, C#,... kinectTCP runs under Win 7, but it frees the programmer from the necessity of using a .net environment.
Dr. Zoran Obradovic has been awarded a new four year grant from Defense Advanced Research Projects Agency. His project is aimed at developing and validating effective predictive modeling technology to achieve challenging sepsis treatment related aims on high dimensional and noisy data at a clinically relevant scale. For more information visit http://www.dabi.temple.edu/dabi/people/zoran/research/darpa_therapy.html
Dr. Slobodan Vucetic and Dr. Zoran Obradovic have been awarded a new three year grant from the National Science Foundation. Their project is aimed at developing a novel discriminative modeling framework for fusion of multi-sensor remote sensing data based on the Gaussian conditional random field model. Project's title is "A Discriminative Modeling Framework for Mining of Spatio-Temporal Data in Remote Sensing" and you are able to find more information about it here.
The Data Analysis and Biomedical Analytics Center extends a warm welcome to Petar Jevtic who joined Prof. Obradovic's lab on August 1st, 2011. Petar Jevtic is a full scholarship, second year PhD student at the Faculty Vilfredo Pareto, University of Turin, Italy. As a visiting scholar at Prof. Obradovic's laboratory his research objectives are to explore the application of stochastic search algorithms in calibration of complex stochastic models in finance and insurance. In conducting his experiments Petar will be using the High Performance Computer available to the center.
Dr. Alexander Yates has been awarded a new one year Faculty Research and Engagement Program grant from the Yahoo Inc. His project is aimed at Domain Adaption for Objection Resolution.
Dr. Latecki has been awarded a new grant from Johnson and Johnson Pharmaceutical Research for the project "Machine Learning Methods for Discovery of Relevant Features"
Dr. Alexander Yates and Dr. Yuhong Guo have been awarded a new four year grant from the National Science Foundation. Their project is aimed at Learning Representations of Language for Domain Adaptation. For more information visit the Project's website
Dr. Latecki has been awarded a new grant from Sandia National Laboratory. His project is aimed atRecovery of 3D Shapes from X-Ray Images.
Dr. Longin Latecki has received 2010 College of Science and Technology Research Excellence Award.
This award was established under the leadership of Dean Hai-Lung Dai to recognize notable recent achievements in research by a faculty member of the College of Science and Technology.
Dr. Latecki has been awarded a new grant from National Science Foundation. His project Perception of Scene Layout by Machines and Visually Impaired Users investigates computational methods for object detection, spatial scene construction, and natural language spatial descriptions derived from real-time visual images to describe prototypical indoor spaces (e.g., rooms, offices, etc.)
Dr. Haibin Ling has been awarded a new two-year grant from the National Science Foundation. His project is aimed at developing a new framework for balancing deformability and discriminability in computer vision.
Dr. Zoran Obradovic is serving as a Co-Chair of the Fourth International Workshop on Mining Multiple Information Sources organized in conjunction with the 10th IEEE International Conference on Data Mining (Sydney, Australia, Dec. 13, 2010).
Papers for this workshop are due by July 23, 2010. For more details see http://www.cse.fau.edu/~xqzhu/mmis/mmis-10/mmis10.html
Dr. Rolf Lakaemper has been awarded a new 3-year grant from National Institute of Standards and Technology. Click here for an overview His project is aimed at creating and experimentally validating a framework by which automated guided vehicles (AGVs), robotic devices that are widely used in factory floors to transport goods, can automatically generate a sufficiently accurate internal map (world model) of its surroundings in order to make them more versatile and useful.
On Dec. 6, 2009, Prof. Obradovic gave a keynote lecture at the 3rd International Workshop on Mining Multiple Information Sources, help in conjunction with IEEE International Conference on Data Mining in Miami, FL.
Topic of his lecure was "Spatio-Temporal Characterization of Aerosols through Active Use of Data from Multiple Sensors."
Following is abstract of his presentation:
One of the main challenges of current climate research is providing Earth-wide characterization of Aerosol Optical Depth (AOD) which indicates the amount of depletion that a beam of radiation undergoes as it passes through the atmosphere. In this talk we will discuss issues related to a statistical approach aimed at better understanding of spatio-temporal distribution of AOD by taking advantage of measurements collected from multiple ground and satellite-based sensors. Challenges to be addressed in contest of global scale AOD estimation include (i) training a predictor for robust performance across multiple accuracy measures; (ii) AOD regression from mixed-distribution spatio-temporal data; (iii) uncertainty analysis of AOD estimation, (iv) active selection of sites for ground based observations, and (v) discovery of major sources of correctable errors in deterministic models.
Dr. Megalooikonomou and Dr. Ling have been awarded a new grant from the National Science Foundation. Their project is aimed at Modeling, Detection, and Analysis of Branching Structures in Medical Imaging.
Dr. Latecki has been awarded a new three year grant from National Science Foundation. His project is aimed at Recovery of 3D Shapes from Single Views.
Dr. Latecki has been awarded a new grant from Air Force Office of Scientific Research (AFOSR). His project is aimed at Robotic Navigation Emulating Human Performance. Project website: http://www.cis.temple.edu/~shape/CG/index.html
Dr. Latecki has been awarded Los Alamos National Lab grant RADIUS: Rapid Automated Decomposition of Images for Ubiquitous Sensing.
Dr. Zoran Obradovic has been named as the recipient of the Temple University Faculty Research Award for 2008-2008.
The official presentation of the award was made at the Faculty Awards Convocation on April 28th 2009 in the Great Court of Mitten Hall.
The following article on Dr. Obradovic's work that has earned him this year's Temple University Research Award appeared in Temple Times of April 17, 2009 http://www.temple.edu/newsroom/2008_2009/04/stories/obradovic.htm (local mirror copy).
In January 2009 Dr. Yuhong Guo has joined the IST Center and the CIS Department as an Assistant Professor. She has just completed a post-doctoral training in machine learning at the Australian National University. Previously, she was a post-doctoral fellow at the University of Alberta where in 2007 she received her Ph.D. while working in the laboratory of Prof. D. Schuurman. Dr. Guo is a recipient of 2005 IJCAI Distinguished Paper Award (1323 papers submitted to this conference, of which 240 were accepted and 3 were honored with this award). Her primary area of research is machine learning and she is also interested in bioinformatics.
A special double issue of Statistical Analysis and Data Mining journal guest edited by Prof. Obradovic and Prof. H. Liu from Arizona State University is published in Dec. 2008 (vol. 2, no. 3-4). This issue contains extended papers of the best studies selected from submissions to the Ninth SIAM International Conference on Data Mining (SDM'09).
On Nov. 04 Dr. Zoran Obradovic, Director of Center for Information Science and Technology, gave a keynote lecture at the IEEE International Conference on Bioinformatics and Biomedicine. This multidisciplinary conference brought together computational scientists from several disciplines and from several continents who exchanged research results in databases, algorithms, interfaces, visualization, modeling, simulation and ontology as applied to high throughput data-rich areas in biology and biomedical engineering.
In his keynote lecture "Functions of Intrinsically Disordered Proteins and Relationship with Human Disease Network," Dr. Obradovic described their award winning predictor of protein disorder (CASP 7) and explained how they recently used it to provide a leap jump in understanding relationship between protein disorder and protein function. In particular, he discussed their characterization of 238 Swiss-Prot functional categories as strongly positively correlated with predicted long intrinsically disordered regions. He also presented the results of their most recent large scale analysis of intrinsic disorder in genes implicated in Human Disease Network. This new study found that intrinsic disorder in disease genes is mainly involved in protein-protein interactions. Genes related to several classes of diseases were found to have significantly higher occurrence of alternative splicing (AS), and strong evidence was provided that intrinsic disorder, together with AS, plays an important role in these classes of diseases.
Haibin Ling whose research interests are in Computer Vision, Medical Imaging, Human Computer Interaction, Machine Learning has joined the CIS Department.
Professor Ling received the B.S. degree in mathematics and the MS degree in computer science from Peking University, China, in 1997 and 2000, respectively, and the PhD degree from the University of Maryland, College Park, in Computer Science in 2006.
From 2000 to 2001, he was an assistant researcher in the Multi-Model User Interface Group at Microsoft Research Asia. From 2006 to 2007, he worked as a postdoctoral scientist at the University of California Los Angeles. After that, he joined Siemens Corporate Research as a research scientist. Since Fall 2008, he has been an Assistant Professor at Temple University.
Dr. Ling's research interests include computer vision, medical image analysis, human computer interaction, and machine learning. He received the Best Student Paper Award at the ACM Symposium on User Interface Software and Technology (UIST) in 2003.
Dr. Latecki has been awarded a new four year grant from National Science Foundation. His project is aimed at Simultaneous Contour Grouping and Medial Axis Estimation. Project website: http://knight.cis.temple.edu/~shape/partshape/contour
Dr. Zoran Obradovic is serving as a Program Chair for 2009 SIAM International Conference on Data Mining (SDM 09) which is one of the premier peer-reviewed forums for sharing research results related to knowledge extraction from large, complex and noisy datasets. Jointly with his program co-chair Prof. Huan Liu from Arizona State University, Prof. Obradovic selected an outstanding expert team of 13 area chairs and 148 program committee members working at academia and industry around 5 continents who will help reviewing manuscripts submitted for SDM07 consideration. Deadline for SDM09 submissions is
Oct. 03, 2008 Oct. 06, 2008, while the conference will be held April 30 - May 2, 2009 at a resort near Reno, Nevada.
The Department of Computer and Information Sciences (CIS) and the Center for Information Science and Technology (IST) at Temple University are recruiting for a tenure-track position at the Assistant Professor level. The position is contingent upon completion of a Ph.D. in computer science or a related field by August 15, 2008. ... READ THE FULL AD ...
Dr. Alexander P. Yates has joined the CIS Department and the IST Center as Assistant Professor. He received his Ph.D. from the University of Washington while working in the laboratory of Dr. Oren Etzioni. Dr. Yates'research interests include computational linguistics and artificial intelligence, specifically information extraction from the Web, entity resolution, natural language interfaces, parsing, machine learning, and probabilistic methods.
The ISTZORAN Group wins the Protein Disorder Prediction category at regions at the seventh Critical Assessments of Structure Prediction experiments meeting (CASP 7), Nov. 26-30, 2006.
The VSL2 predictor of intrinsically disordered protein regions developed by the ISTZORAN group has been rated as the best model at the 7th Critical Assessments of Structure Prediction experiments meeting (CASP 7). An older version of this model was the best model at CASP6.
You can try the VSL2 predictor or download the VSL2 predictor package and set it up on your own machine (detailed instructions). You can choose whether to use computationally demanding features derived from PSI-BLAST profiles and/or secondary structure predictions.
Citation: Peng K., Radivojac P., Vucetic S., Dunker A.K., and Obradovic Z., Length-Dependent Prediction of Protein Intrinsic Disorder, BMC Bioinformatics 7:208, 2006.
Dr. Slobodan Vucetic received CAREER award from NSF. CAREER: Memory-Constrained Predictive Data Mining
Dr. Slobodan Vucetic received CAREER award from National Science Foundation. CAREER: Memory-Constrained Predictive Data Mining
September 2005 marked five years from inauguration of Information Science and Technology (IST) Center at Temple University. The main mission of the IST Center is advanced research and education aimed toward solving challenging problems in data mining, machine learning, multimedia databases, data compression, biomedical informatics, pattern recognition, computer vision, robot mapping, computational genomics, and artificial intelligence. The results of investigations at the IST Center in 2000-2005 period were published as about 60 journal papers, 123 refereed conference papers, and 9 refereed book chapters. The awarded research funding for projects that involve investigators from IST Center was near 6 million dollars. Specific activities at the IST Center of the first five year period are summarized in this report. Read the full report: HTML or PDF