PTM probabilistic topic mapping model for mining parallel document collections.

Medical Case%2Dbased Retrieval by Leveraging Medical Ontology and Physician Feedback UIUC%2DIBM at ImageCLEF 2010.
Abstract: This paper reports the experiment results of the UIUC-IBM team in participating in the medical case retrieval task of ImageCLEF 2010. We experimented with multiple methods to leverage medical ontology and user (physician) feedback; both have worked very well, achieving the best retrieval performance among all the submissions.
Localized Supervised Metric Learning on Temporal Physiological Data.

Social action tracking via noise tolerant time%2Dvarying factor graphs.
Users’behaviors(actions)inasocialnetworkareinfluencedbyvarious factors such as personal interests, social influence, and global trends. However, few publications systematicallystudy how social actions evolve in a dynamic social network and towhat extent different factors affect the user actions. In this paper, we propose a Noise Tolerant Time-varying Factor Graph Model (NTT-FGM) for modeling and predicting social actions. NTT-FGM simultaneously models social network structure, user attributes and user action history for better prediction of the users ’ future actions. More specifically, a user’s action at time t is generated by her latent state at t, which is influenced by her attributes,herownlatentstateattimet−1andherneighbors’ states attimetandt−1. Basedonthisintuition,weformalizethe social action tracking problem using the NTT-FGM model; then present an efficient algorithm to learn the model, by combining the ideas from both continuous linear system and Markov random field. Finally, we present a case study of our model on predicting future social actions. We validate the model on three different types ofreal-worlddatasets. Qualitatively,ourmodelcandiscover interestingpatternsofthesocialdynamics. Quantitatively,experimental resultsshowthattheproposedmethodoutperformsseveralbaseline methods for social actionprediction. Categories andSubject Descriptors
Temporal recommendation on graphs via long%2D and short%2Dterm preference fusion.

ContexTour Contextual Contour Analysis on Dynamic Multi%2Drelational Clustering.

ContexTour Contextual Contour Analysis on Dynamic Multi%2Drelational Clustering.

FacetAtlas Multifaceted Visualization for Rich Text Corpora.

SmallBlue Social Network Analysis for Expertise Search and Collective Intelligence.

Two Heads Better Than One Metric Active Learning and its Applications for IT Service Classification.

MetaFac community discovery via relational hypergraph factorization.

Social influence analysis in large%2Dscale networks.

HiMap Adaptive visualization of large%2Dscale online social networks.
Visualizing large-scale online social network is a challenging yet essential task. This paper presents HiMap, a system that visualizes it by clustered graph via hierarchical grouping and summarization. HiMap employs a novel adaptive data loading technique to accurately control the visual density of each graph view, and along with the optimized layout algorithm and the two kinds of edge bundling methods, to effectively avoid the visual clutter commonly found in previous social network visualization tools. HiMap also provides an integrated suite of interactions to allow the users to easily navigate the social map with smooth and coherent view transitions to keep their momentum. Finally, we confirm the effectiveness of HiMap algorithms through graph-travesal based evaluations.
MultiVis Content%2DBased Social Network Exploration through Multi%2Dway Visual Analysis.

Extracting community structure through relational hypergraphs.

Clustered subset selection and its applications on it service metrics.

Two Heads Better Than One Pattern Discovery in Time%2DEvolving Multi%2Daspect Data.

Hierarchical%2C Parameter%2DFree Community Discovery.
Abstract. Given a large bipartite graph (like document-term, or userproduct graph), how can we find meaningful communities, quickly, and automatically? We propose to look for community hierarchies, with communities-within-communities. Our proposed method, the Context-specific Cluster Tree (CCT) finds such communities at multiple levels, with no user intervention, based on information theoretic principles (MDL). More specifically, it partitions the graph into progressively more refined subgraphs, allowing users to quickly navigate from the global, coarse structure of a graph to more focused and local patterns. As a fringe benefit, and also as an additional indication of its quality, it also achieves better compression than typical, non-hierarchical methods. We demonstrate its scalability and effectiveness on real, large graphs. 1
Scalable Tensor Decompositions for Multi%2Daspect Data Mining.

DisCo Distributed Co%2Dclustering with Map%2DReduce A Case Study towards Petabyte%2DScale End%2Dto%2DEnd Mining.
Huge datasets are becoming prevalent; even as researchers, we now routinely have to work with datasets that are up to a few terabytes in size. Interesting real-world applications produce huge volumes of messy data. The mining process involves several steps, starting from pre-processing the raw data to estimating the final models. As data become more abundant, scalable and easyto-use tools for distributed processing are also emerging. Among those, Map-Reduce has been widely embraced by both academia and industry. In database terms, Map-Reduce is a simple yet powerful execution engine, which can be complemented with other data storage and management components, as necessary. In this paper we describe our experiences and findings in applying Map-Reduce, from raw data to final models, on an important mining task. In particular, we focus on co-clustering, which has been studied in many applications such as text mining, collaborative filtering, bio-informatics, graph mining. We propose the Distributed Co-clustering (DisCo) framework, which introduces practical approaches for distributed data pre-processing, and co-clustering. We develop DisCo using Hadoop, an open source Map-Reduce implementation. We show that DisCo can scale well and efficiently process and analyze extremely large datasets (up to several hundreds of gigabytes) on commodity hardware. 1
SCOOP Automated Social Recommendation in Enterprise Process Management.

Bayesian tensor analysis.

Two heads better than one pattern discovery in time%2Devolving multi%2Daspect data.

Colibri fast mining of large static and dynamic graphs.
Low-rank approximations of the adjacency matrix of a graph are essential in finding patterns (such as communities) and detecting anomalies. Additionally, it is desirable to track the low-rank structure as the graph evolves over time, efficiently and within limited storage. Real graphs typically have thousands or millions of nodes, but are usually very sparse. However, standard decompositions such as SVD do not preserve sparsity. This has led to the development of methods such as CUR and CMD, which seek a nonorthogonal basis by sampling the columns and/or rows of the sparse matrix. However, these approaches will typically produce overcomplete bases, which wastes both space and time. In this paper we propose the family of Colibri methods to deal with these challenges. Our version for static graphs, Colibri-S, iteratively finds a nonredundant basis and we prove that it has no loss of accuracy compared to the best competitors (CUR and CMD), while achieving significant savings in space and time: on real data, Colibri-S requires much less space and is orders of magnitude faster (in proportion to the square of the number of non-redundant columns). Additionally, we propose an efficient update algorithm for dynamic, time-evolving graphs, Colibri-D. Our evaluation on a large, real network traffic dataset shows that Colibri-D is over 100 times faster than the best published competitor (CMD).
Two heads better than one pattern discovery in time%2Devolving multi%2Daspect data.

Bayesian Tensor Approach for 3%2DD Face Modeling.

Incremental pattern discovery on streams%2C graphs and tensors.
0326322 IIS-0534205 and under the auspices of the U.S. Department of Energy by Lawrence Livermore
Less is More Sparse Graph Mining with Compact Matrix Decomposition.

Incremental tensor analysis Theory and applications.

Hiding in the Crowd Privacy Preservation on Evolving Streams through Correlation Tracking.
We address the problem of preserving privacy in streams, which has received surprisingly limited attention. For static data, a well-studied and widely used approach is based on random perturbation of the data values. However, streams pose additional challenges. First, analysis of the data has to be performed incrementally, using limited processing time and buffer space, making batch approaches unsuitable. Second, the characteristics of streams evolve over time. Consequently, approaches based on global analysis of the data are not adequate. We show that it is possible to efficiently and effectively track the correlation and autocorrelation structure of multivariate streams and leverage it to add noise which maximally preserves privacy, in the sense that it is very hard to remove. Our techniques achieve much better results than previous static, global approaches, while requiring limited processing time and memory. We provide both a mathematical analysis and experimental evaluation on real data to validate the correctness, efficiency, and effectiveness of our algorithms. 1.
Dynamic Mixture Models for Multiple Time%2DSeries.
Traditional probabilistic mixture models such as Latent Dirichlet Allocation imply that data records (such as documents) are fully exchangeable. However, data are naturally collected along time, thus obey some order in time. In this paper, we present Dynamic Mixture Models (DMMs) for online pattern discovery in multiple time series. DMMs do not have the noticeable drawback of the SVD-based methods for data streams: negative values in hidden variables are often produced even with all nonnegative inputs. We apply DMM models to two real-world datasets, and achieve significantly better results with intuitive interpretation. 1
GraphScope parameter%2Dfree mining of large time%2Devolving graphs.

Less is More Compact Matrix Decomposition for Large Sparse Graphs.
Given a large sparse graph, how can we find patterns and anomalies? Several important applications can be modeled as large sparse graphs, e.g., network traffic monitoring, research citation network analysis, social network analysis, and regulatory networks in genes. Low rank decompositions, such as SVD and CUR, are powerful techniques for revealing latent/hidden variables and associated patterns from high dimensional data. However, those methods often ignore the sparsity property of the graph, and hence usually incur too high memory and computational cost to be practical. We propose a novel method, the Compact Matrix Decomposition (CMD), to compute sparse low rank approximations. CMD dramatically reduces both the computation cost and the space requirements over existing decomposition methods (SVD, CUR). Using CMD as the key building block, we further propose procedures to efficiently construct and analyze dynamic graphs from real-time application data. We provide theoretical guarantee for our methods, and present results on two real, large datasets, one on network flow data (100GB trace of 22K hosts over one month) and one on DBLP (200MB over 25 years). We show that CMD is often an order of magnitude more efficient than the state of the art (SVD and CUR): it is over 10X faster, but requires less than 1/10 of the space, for the same reconstruction accuracy. Finally, we demonstrate how CMD is used for detecting anomalies and monitoring timeevolving graphs, in which it successfully detects worm-like hierarchical scanning patterns in real network data. 1
Mining large graphs and streams using matrix and tensor tools.

Intelligent system monitoring on large clusters.
InteMon is a prototype monitoring and mining system for large clusters. Currently, it monitors over 100 hosts of a prototype data center at CMU. It uses the SNMP protocol and it stores the monitoring data in an mySQL database. Then, it allows for visualization of the time-series data using a JSP web-based frontend interface for users. What sets it apart from other cluster monitoring systems is its ability to automatically analyze the monitoring data in real time and alert the users for potential anomalies. It uses state of the art stream mining methods, it has a sophisticated definition of anomalies (broken correlations among input streams), and it can also pinpoint the reason of the anomaly. InteMon has a user-friendly GUI, it allows the users to perform interactive mining tasks, and it is fully operational. 1.
Window%2Dbased Tensor Analysis on High%2Ddimensional and Multi%2Daspect Streams.
Data stream values are often associated with multiple aspects. For example, each value from environmental sensors may have an associated type (e.g., temperature, humidity, etc) as well as location. Aside from timestamp, type and location are the two additional aspects. How to model such streams? How to simultaneously find patterns within and across the multiple aspects? How to do it incrementally in a streaming fashion? In this paper, all these problems are addressed through a general data model, tensor streams, and an effective algorithmic framework, window-based tensor analysis (WTA). Two variations of WTA, independentwindow tensor analysis (IW) and moving-window tensor analysis (MW), are presented and evaluated extensively on real datasets. Finally, we illustrate one important application, Multi-Aspect Correlation Analysis (MACA), which uses WTA and we demonstrate its effectiveness on an environmental monitoring application. 1
Local Correlation Tracking in Time Series.
We address the problem of capturing and tracking local correlations among time evolving time series. Our approach is based on comparing the local auto-covariance matrices (via their spectral decompositions) of each series and generalizes the notion of linear cross-correlation. In this way, it is possible to concisely capture a wide variety of local patterns or trends. Our method produces a general similarity score, which evolves over time, and accurately reflects the changing relationships. Finally, it can also be estimated incrementally, in a streaming setting. We demonstrate its usefulness, robustness and efficiency on a wide range of real datasets. 1
Distributed Pattern Discovery in Multiple Streams.
and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation, or other funding parties.
Beyond streams and graphs dynamic tensor analysis.
How do we find patterns in author-keyword associations, evolving over time? Or in DataCubes, with product-branchcustomer sales information? Matrix decompositions, like principal component analysis (PCA) and variants, are invaluable tools for mining, dimensionality reduction, feature selection, rule identification in numerous settings like streaming data, text, graphs, social networks and many more. However, they have only two orders, like author and keyword, in the above example. We propose to envision such higher order data as tensors, and tap the vast literature on the topic. However, these methods do not necessarily scale up, let alone operate on semi-infinite streams. Thus, we introduce the dynamic tensor analysis (DTA) method, and its variants. DTA provides a compact summary for high-order and high-dimensional data, and it also reveals the hidden correlations. Algorithmically, we designed DTA very carefully so that it is (a) scalable, (b) space efficient (it does not need to store the past) and (c) fully automatic with no need for user defined parameters. Moreover, we propose STA, a streaming tensor analysis method, which provides a fast, streaming approximation to DTA. We implemented all our methods, and applied them in two real settings, namely, anomaly detection and multi-way latent semantic indexing. We used two real, large datasets, one on network flow data (100GB over 1 month) and one from DBLP (200MB over 25 years). Our experiments show that our methods are fast, accurate and that they find interesting patterns and outliers on the real datasets. 1.
Distributed Pattern Discovery in Multiple Streams.
and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation, or other funding parties.
InteMon Intelligent System Monitoring on Large Clusters.
InteMon is a prototype monitoring and mining system for large clusters. Currently, it monitors over 100 hosts of a prototype data center at CMU. It uses the SNMP protocol and it stores the monitoring data in an mySQL database. Then, it allows for visualization of the time-series data using a JSP web-based frontend interface for users. What sets it apart from other cluster monitoring systems is its ability to automatically analyze the monitoring data in real time and alert the users for potential anomalies. It uses state of the art stream mining methods, it has a sophisticated definition of anomalies (broken correlations among input streams), and it can also pinpoint the reason of the anomaly. InteMon has a user-friendly GUI, it allows the users to perform interactive mining tasks, and it is fully operational. 1.
Spatio%2Dtemporal join selectivity.
Moving objects databases managing spatial objects with continuously changing position and extent over time have recently found large interest in the database community. Queries about moving objects become particularly interesting when they ask for temporal changes in the topological relationships between evolving spatial objects. A concept of spatio-temporal predicates has been proposed to describe these relationships. The goal of this paper is to design efficient algorithms for them so that they can be used in spatiotemporal joins and selections. This paper proposes not to design an algorithm for each new predicate individually but to employ a generic algorithmic scheme which is able to cover present and future predicate definitions. 1.
InteMon continuous mining of sensor data in large%2Dscale self%2Dinfrastructures.
Modern data centers have a large number of components that must be monitored, including servers, switches/routers, and environmental control systems. This paper describes InteMon, a prototype monitoring and mining system for data centers. It uses the SNMP protocol to monitor a new data center at Carnegie Mellon. It stores the monitoring data in a MySQL database, allowing visualization of the time-series data using a JSP web-based frontend interface for system administrators. What sets InteMon apart from other cluster monitoring systems is its ability to automatically analyze correlations in the monitoring data in real time and alert administrators of potential anomalies. It uses efficient, state of the art stream mining methods to report broken correlations among input streams. It also uses these methods to intelligently compress historical data and avoid the need for administrators to configure threshold-based monitoring bands.
Online Latent Variable Detection in Sensor Networks.

Neighborhood Formation and Anomaly Detection in Bipartite Graphs.
Many real applications can be modeled using bipartite graphs, such as users vs. files in a P2P system, traders vs. stocks in a financial trading system, conferences vs. authors in a scientific publication network, and so on. We introduce two operations on bipartite graphs: 1) identifying similar nodes (Neighborhood formation), and 2) finding abnormal nodes (Anomaly detection). And we propose algorithms to compute the neighborhood for each node using random walk with restarts and graph partitioning; we also propose algorithms to identify abnormal nodes, using neighborhood information. We evaluate the quality of neighborhoods based on semantics of the datasets, and we also measure the performance of the anomaly detection algorithm with manually injected anomalies. Both effectiveness and efficiency of the methods are confirmed by experiments on several real datasets. 1
Streaming Pattern Discovery in Multiple Time%2DSeries.
In this paper, we introduce SPIRIT (Streaming  Pattern dIscoveRy in multIple Timeseries)  . Given n numerical data streams, all  of whose values we observe at each time tick  t, SPIRIT can incrementally find correlations  and hidden variables, which summarise the  key trends in the entire stream collection.
Relevance search and anomaly detection in bipartite graphs.
Many real applications can be modeled using bipartite graphs, such as users vs. files in a P2P system, traders vs. stocks in a financial trading system, conferences vs. authors in a scientific publication network, and so on. We introduce two operations on bipartite graphs: 1) identifying similar nodes (relevance search), and 2) finding nodes connecting irrelevant nodes (anomaly detection). And we propose algorithms to compute the relevance score for each node using random walk with restarts and graph partitioning; we also propose algorithms to identify anomalies, using relevance scores. We evaluate the quality of relevance search based on semantics of the datasets, and we also measure the performance of the anomaly detection algorithm with manually injected anomalies. Both effectiveness and efficiency of the methods are confirmed by experiments on several real datasets. 1.
Querying about the Past%2C the Present%2C and the Future in Spatio%2DTemporal.
Moving objects (e.g., vehicles in road networks) continuously generate large amounts of spatio-temporal information in the form of data streams. Efficient management of such streams is a challenging goal due to the highly dynamic nature of the data and the need for fast, on-line computations. In this paper we present a novel approach for approximate query processing about the present, past, or the future in spatio-temporal databases. In particular, we first propose an incrementally updateable, multi-dimensional histogram for present-time queries. Second, we develop a general architecture for maintaining and querying historical data. Third, we implement a stochastic approach for predicting the results of queries that refer to the future. Finally, we experimentally prove the effectiveness and efficiency of our techniques using a realistic simulation. 1.
Querying about the Past%2C the Present%2C and the Future in Spatio%2DTemporal.
Moving objects (e.g., vehicles in road networks) continuously generate large amounts of spatio-temporal information in the form of data streams. Efficient management of such streams is a challenging goal due to the highly dynamic nature of the data and the need for fast, on-line computations. In this paper we present a novel approach for approximate query processing about the present, past, or the future in spatio-temporal databases. In particular, we first propose an incrementally updateable, multi-dimensional histogram for present-time queries. Second, we develop a general architecture for maintaining and querying historical data. Third, we implement a stochastic approach for predicting the results of queries that refer to the future. Finally, we experimentally prove the effectiveness and efficiency of our techniques using a realistic simulation. 1.
Selectivity Estimation for Predictive Spatio%2DTemporal Queries.
This paper proposes a cost model for selectivity estimation of predictive spatio-temporal window queries. Initially, we focus on uniform data proposing formulae that capture both points and rectangles, and any type of object/query mobility combination (i.e., dynamic objects, dynamic queries or both). Then, we apply the model to non-uniform datasets by introducing spatio-temporal histograms, which in addition to the spatial, also consider the velocity distributions during partitioning. The advantages of our techniques are (i) high accuracy (1-2 orders of magnitude lower error than previous techniques), (ii) ability to handle all query types, and (iii) efficient handling of updates.
The TPR%2A%2DTree An Optimized Spatio%2DTemporal Access Method for Predictive Queries.
A predictive spatio-temporal query retrieves the  set of moving objects that will intersect a query  window during a future time interval. Currently,  the only access method for processing such  queries in practice is the TPR-tree. In this paper  we first perform an analysis to determine the  factors that affect the performance of predictive  queries and show that several of these factors are  not considered by the TPR-tree, which uses the  insertion/deletion algorithms of the R*-tree  designed for static data. Motivated by this, we  propose a new index structure called the TPR*-  tree, which takes into account the unique features  of dynamic objects through a set of improved  construction algorithms. In addition, we provide  cost models that determine the optimal  performance achievable by any data-partition  spatio-temporal access method. Using  experimental comparison, we illustrate that the  TPR*-tree is nearly-optimal and significantly  outperforms the TPR-tree under all conditions.
Analysis of predictive spatio%2Dtemporal queries.
this paper we present probabilistic cost models that estimate the selectivity of spatio-temporal window queries and joins, and the expected distance between a query and its nearest neighbor(s). Our models capture any query/object mobility combination (moving queries, moving objects or both) and any data type (points and rectangles) in arbitrary dimensionality. In addition, we develop specialized spatio-temporal histograms, which take into account both location and velocity information, and can be incrementally maintained. Extensive performance evaluation verifies that the proposed techniques produce highly accurate estimation on both uniform and non-uniform data
Indexing and Retrieval of Historical Aggregate Information about Moving Objects.
Spatio-temporal databases store information about the positions of individual objects over time. In many applications however, such as traffic supervision or mobile communication systems, only summarized data, like the average number of cars in an area for a specific period, or phones serviced by a cell each day, is required. Although this information can be obtained from operational databases, its computation is expensive, rendering online processing inapplicable. A vital solution is the construction of a spatiotemporal data warehouse. In this paper, we describe a framework for supporting OLAP operations over spatiotemporal data. We argue that the spatial and temporal dimensions should be modeled as a combined dimension on the data cube and present data structures, which integrate spatiotemporal indexing with pre-aggregation. While the well-known materialization techniques require a-priori knowledge of the grouping hierarchy, we develop methods that utilize the proposed structures for efficient execution of ad-hoc group-bys. Our techniques can be used for both static and dynamic dimensions. 1
