Papers.

Give-to-Get : free-riding resilient video-on-demand in P2P systems

Enforcing Transitivity in Coreference Resolution

El Poder de buscar

El crimen de lord Arthur Savile: y otros relatos

Efficient, Feature-based, Conditional Random Field Parsing

Different roles of pitch and duration in distinguishing word stress in English

Developing a Distributed Electronic Health-Record Store for India

Complex networks in recommendation systems

Code Generation Strategies from AADL Architectural Descriptions Targeting the High Integrity Domain

Code Generation Strategies for Partitioned Systems

Design and verification of Distributed Real-time Embed-ded (DRE) systems are crucial because a failure or a secu-rity problem may cause loss of life or the premature end of a mission. Code for such systems must be verified to avoid failures and errors. To improve dependability and reliability, the concept of partitioned architecture has been proposed. Partitioned architectures isolate partitions and reduce failures propagation. However, system’s configura-tion as well as application code remain hand-written, which is error-prone, difficult to check and certify. In this paper, we present an approach to automatically generate and configure partitioned systems. We model par-titioned systems using a language suitable for real-time em-bedded systems. Then, we automatically generate code from models. Generated code follows the semantics of the model so it enforces specified requirements. 1

BarterCast : Fully Distributed Sharing-Ratio Enforcement in BitTorrent

Approximating Learning Curves for Active-Learning-Driven Annotation.

Applying computational semantics to the real-time communication of skill knowledge

Agent Organized Networks Redux

Agent-based sensor coalition formation

A Small-Vocabulary Shared Task for Medical Speech Translation

A Simple Publishing Interface For Learning Object Repositories

A Legal Geography of Yugoslavia’s Disintegration

5. Argument Small Clauses

Psychological backgrounds for inducing cooperation in peer-to-peer television

Load-balancing and caching for collection selection architectures

The End of an Architectural Era (It's Time for a Complete Rewrite)

The H-Store paper: argues that the traditional RDBMS architecture is obsolete and that specialized engines will outperform general-purpose 'one size fits all' databases by a large factor.

Annotating Music Collections: How Content-Based Similarity Helps To Propagate Labels.

TEXT ANALYTICS AND DATA ACCESS AS SERVICES - A Case Study in Transforming a Legacy Client-server Text Analytics Workbench and Framework to SOA

As business information is made available via the intranet and Internet, there is a growing need to quickly analyze the resulting mountain of information to infer business insights. For instance, analyzing a company's patent database against another's to find the patents that are cross-licensable. IBM Research's Business Insight Workbench (BIW) is a text mining and analytics tool that allows end-users to explore, understand, and analyze business information in order to come up with such insight. However, the first incarnation of BIWused a thick- client architecture with a database back-end. While very successful, the architecture caused limitations in the tool's flexibility, scalability, and deployment. In this paper we discuss our initial experiences in converting BIW into a modern Service-Oriented Architecture. We also provide some insights into our design choices and also outline some lessons learned.

Load-Balancing and Caching for Collection Selection Architectures

A Reappraisal of Sentence and Token Splitting for Life Sciences Documents.

Learning Alignments and Leveraging Natural Logic

We describe an approach to textual inference that improves alignments at both the typed dependency level and at a deeper semantic level. We present a machine learning approach to alignment scoring, a stochastic search procedure, and a new tool that finds deeper semantic alignments, allowing rapid development of semantic features over the aligned graphs. Further, we describe a complementary semantic component based on natural logic, which shows an added gain of 3.13% accuracy on the RTE3 test set.

Mutually unbiased bases and orthogonal decompositions of Lie algebras

We establish a connection between the problem of constructing maximal collections of mutually unbiased bases (MUBs) and an open problem in the theory of Lie algebras. More precisely, we show that a collection of µ MUBs in n gives rise to a collection of µ Cartan subalgebras of the special linear Lie algebra sln() that are pairwise orthogonal with respect to the Killing form, where = or =. In particular, a complete collection of MUBs in n gives rise to a so-called orthogonal decomposition (OD) of sln(). The converse holds if the Cartan subalgebras in the OD are also †-closed, i.e., closed under the adjoint operation. In this case, the Cartan subalgebras have unitary bases, and the above correspondence becomes equivalent to a result of [2] relating collections of MUBs to collections of maximal commuting classes of unitary error bases, i.e., orthogonal unitary matrices. This connection implies that for n ≤ 5 an essentially unique complete collection of MUBs exists. We define monomial MUBs, a class of which all known MUB constructions are members, and use the above connection to show that for n = 6 there are at most three monomial MUBs.

Geolocation of RF Emitters by Many UAVs

Advances in Technology of KODAK NEXPRESS Digital Production Color Presses

The stochastic screen has traditionally been used to preserve images with fine features as well as to reduce printing Moiré of images that contain input periodic structures. However, the stochastic screen also has experienced more difficulty in maintaining color stability in long press runs. Kodak, using its multilevel printing architecture, has created a new stochastic screening technology, with long-run color stability comparable to that of traditional screens. The new screening process has a unique AM and FM structure consisting of variable dot size, dot shape, dot density, and dot positions. The structure simulates the receptive field structure of the human retina to achieve the desired image smoothness. Kodak's high information capacity, multilevel printing technology in conjunction with nonlinear LED exposure, can further be expanded to production printing with even finer contone-screen printing. This paper will discuss the advances of these technologies.

Perception of disfluency: language differences and listener bias

The Certainty of chance: Review of Bod, R., Hay, J. and Jannedy, S. (2003) Probabiltic Linguistics

Visual-shift adaptation is composed of separable sensory and task-dependent effects

Decomposes prism/visual-shift adaptation into a sensory realignment component and a task-dependent component, showing the two adapt with different dynamics and generalisation.

Ricardo Baeza-Yates speaks out

Where were we

Locating RF emitters with large UAV teams

On the value of temporal information in information retrieval

Structured co-evolution of models and web application platforms

Swashup

Distributed programming has shifted from private networks to the Internet using heterogeneous Web APIs. This enables the creation of situational applications of composed services exposing user interfaces, i.e., mashups. However, this programmable Web lacks unified models that can facilitate mashup creation, reuse, and deployments. This poster demonstrates a platform to facilitate Web 2.0 mashups.

Confessions of a used programming language salesman

For many years I had been fruitlessly trying to sell functional programming and Haskell to solve real world problems such as scripting and data-intensive three-tier distributed web applications. The lack of widespread adoption of Haskell is a real pity. Functional programming concepts are key to curing many of the headaches that plague the majority of programmers, who today are forced to use imperative languages. If the mountain won't come to Mohammed, Mohammed must go to the mountain, and so I left academia to join industry. Instead of trying to convince imperative programmers to forget everything they already know and learn something completely new, I decided to infuse existing imperative object-oriented programming languages with functional programming features. As a result, functional programming has finally reached the masses, except that it is called Visual Basic 9 instead of Haskell 98.

Lost in translation: formalizing proposed extensions to C#

Current real-world software applications typically involve heavy use of relational and XML data and their query languages. Unfortunately object-oriented languages and database query languages are based on different semantic foundations and optimization strategies. The resulting ''ROX (Relations, Objects, XML) impedance mismatc'' makes life very difficult for developers.

Adaptive inlining and on-stack replacement in the CACAO virtual machine

Cost-effective outbreak detection in networks

Given a water distribution network, where should we place sensors toquickly detect contaminants? Or, which blogs should we read to avoid missing important stories?. These seemingly different problems share common structure: Outbreak detection can be modeled as selecting nodes (sensor locations, blogs) in a network, in order to detect the spreading of a virus or information asquickly as possible. We present a general methodology for near optimal sensor placement in these and related problems. We demonstrate that many realistic outbreak detection objectives (e.g., detection likelihood, population affected) exhibit the property of "submodularity". We exploit submodularity to develop an efficient algorithm that scales to large problems, achieving near optimal placements, while being 700 times faster than a simple greedy algorithm. We also derive online bounds on the quality of the placements obtained by any algorithm. Our algorithms and bounds also handle cases where nodes (sensor locations, blogs) have different costs. We evaluate our approach on several large real-world problems,including a model of a water distribution network from the EPA, andreal blog data. The obtained sensor placements are provably near optimal, providing a constant fraction of the optimal solution. We show that the approach scales, achieving speedups and savings in storage of several orders of magnitude. We also show how the approach leads to deeper insights in both applications, answering multicriteria trade-off, cost-sensitivity and generalization questions.

Extracting semantic relations from query logs

Search results using timeline visualizations

The impact of caching on search engines

Nonmyopic active learning of Gaussian processes: an exploration-exploitation approach

Regularly annotated set constraints

Handling heterogeneous data sources in a SOA environment with service data objects (SDO)

A Co-Plot analysis of logs and models of parallel workloads

We present a multivariate analysis technique called Co-Plot that is especially suitable for few samples of many variables. Co-Plot embeds the multidimensional samples in two dimensions, in a way that allows key variables to be identified, and relations between both variables and observations to be analyzed together. When applied to the workloads on parallel supercomputers, we find two stable perpendicular axes of highly correlated variables, one representing individual job attributes and the other representing multijob attributes. The different workloads, on the other hand, are rather different from one another, and may also change over time. Synthetic models for workload generation are also analyzed, and found to be reasonable in the sense that they span the same range of variable combinations as the real workloads. However, the spread of real workloads implies that a single model cannot be similar to all of them. This leads us to construct a parameterized model, with parameters that correspond to the two axes identified above. We also find that existing models do not model the temporal structure of the workload well, and hence are wanting for tasks such as comparing schedulers, and that the common methodology for load manipulation of workloads is problematic.