Abstract

espite the advancement in the information extraction area, the task of identifying associated relations among named entities within a text document remains a significant challenge. Existing automated approaches lack human precision and they also struggle to handle erroneous documents. In this paper, we propose a crowdsourcing-based approach to improve the accuracy of the generated relations from the existing extraction techniques. Our idea is to gather judgments on the extracted relations of an article from the interested users. By contributing, the users in return remember the facts related to a document. This paper presents the complete design of the approach along with a user study done with twelve participants. Results show that the users rated the proposed system positively and were willing to contribute their time and energy for the task.

Abstract

A set of tools that demostrate the concept of a re-narration web, that aims to add alternative, target specific narrations of web documents as an extension of the web protocol with the intent of assisting a process of content dissemination and access that is inclusive of non-literate people.

Abstract

We propose Alipi, a distributed and participatory approach for re-narrating web pages for the purpose of rendering the content more accessible. This model supports alternative descriptions for a web page or parts of it via rewriting or re-narration for a given target audience by volunteers. The goal is to render the Web accessible to people across varied abilities, age, economic situation, language and geographic locations. We present the motivation, architecture and prototype implementation of Alipi.

Abstract

Workflows are ubiquitous in educational ERP and process management. The question we address in this paper is how to specify and design verifiable workflows for educational processes. To specify workflows, we borrow a simple algebraic notation from computer science. We illustrate the use of this notation through a series of typical workflow examples. We show how an algebraic specification is more expressive than a graphical specification. We then show how these algebraic specifications may be represented in the Scribble specification language, which has in built tools for verification.

Abstract

Content creation and management is an inevitable part of teaching a course. This paper describes a novel way of handling this problem using Org-mode, a recently created text based information management tool being used within the Emacs user community. We list certain desirable features, specially for the purposes of computer science courses, such as support for collaborative development and literate programming. We show how Org-mode compares favourably over other approaches like wikis and other content and course management systems. We describe why the combination of Org-mode and version control is suitable for creating and publishing content quickly, with minimum overhead in a collaborative manner.

Abstract

The rapid release model of software introduces frequent updates to the existing software every twelve-eighteen weeks, forcing a user to get accustomed to its new features. We propose an experimental study to compare the learning gaps that are introduced when a user is exposed to a new version of a software, with which he is previously acquainted. In order to explore the problem, we propose four models of a machine, with each model involving an update either to the functionality, or to the user interface, or both. We conducted a between-subjects experimental study with thirty-two participants who performed two tasks successively on two models of a machine, the second one being a updated model of the first. The analysis of the data using ANOVA implies that a change in the user interface dominates a change in the functionality. Results indicate that 88% of the errors were caused due to a change in the user interface. 87.5% of the users who underwent a change in the user interface hold this change responsible for the learning gap, while only 56.25% users who underwent a change in the functionality consider it to be a potential reason for the learning gap.

Abstract

The astonishing diversity in folding patterns of RNA three-dimensional (3D) structures is crafted by myriads of noncovalent contacts, of which base pairing and stacking are the most prominent. A systematic and comprehensive classification and annotation of these interactions is necessary for a molecular-level understanding of their roles. However, unlike in the case of base pairing, where a widely accepted nomenclature and classification scheme exists in the public domain, currently available classification schemes for base–base stacking need major enhancements to comprehensively capture the necessary features underlying the rich stacking diversity in RNA. Here, we extend the previous stacking classification based on nucleobase interacting faces by introducing a structurally intuitive geometry-cum topology-based scheme. Specifically, a stack is first classified in terms of the geometry described by the relative orientation of the glycosidic bonds, which generates eight basic stacking geometric families for heterodimeric stacks and six of those for homodimeric stacks. Further annotation in terms of the identity of the bases and the region of involvement of purines (five-membered, six-membered, or both rings) leads to the enumeration of 384 distinct RNA base stacks. Based on our classification scheme, we present an algorithm for automated identification of stacks in RNA crystal structures and analyze the stacking context in selected RNA structures. Overall, the work described here is expected to greatly facilitate the structure-based RNA research.

Abstract

The AXES project participated in the interactive known-item search task (KIS) and the interactive instance search task (INS) for TRECVid 2011. We used the same system architecture and a nearly identical user interface for both the KIS and INS tasks. Both systems made use of text search on ASR, visual concept detectors, and visual similarity search. The user experiments were carried out with media professionals and media students at the Netherlands Institute for Sound and Vision, with media professionals performing the KIS task and media students participating in the INS task. This paper describes the results and findings of our experiments.

Abstract

Advances in cameras and web technology have madeit easy to capture and share large amounts of video data over toa large number of people. A large number of cameras overseepublic and semi-public spaces today. These raise concerns on theunintentional and unwarranted invasion of the privacy of individ-uals caught in the videos. To address these concerns, automatedmethods tode-identifyindividuals in these videos are necessary.De-identification does not aim at destroying all informationinvolving the individuals. Its ideal goals are to obscure the identityof the actor without obscuring the action. This paper outlinesthe scenarios in which de-identification is required and the issuesbrought out by those. We also present an approach to de-identifyindividuals from videos. Our approach involves tracking andsegmenting individuals in a conservative voxel space involvingx, y,and time. A de-identification transformation is applied perframe using these voxels to obscure the identity. Face, silhouette,gait, and other characteristics need to be obscured, ideally. Weshow results of our scheme on a number of videos and forseveral variations of the transformations. We present the resultsof applying algorithmic identification on the transformed videos.We also present the results of a user-study to evaluate how wellhumans can identify individuals from the transformed videos

Abstract

Raytracing dynamic scenes at interactive rates have received a lot of attention recently. We present a few strategies forhigh performance raytracing on a commodity GPU. The construction of grids needs sorting, which is fast on today’s GPUs. The grid isthus the acceleration structure of choice for dynamic scenes as per-frame rebuilding is required. We advocate the use of appropriatedata structures for each stage of raytracing, resulting in multiple structure building per frame. A perspective grid built for the cameraachieves perfect coherence for primary rays. A perspective grid built with respect to each light source provides the best performancefor shadow rays. Spherical grids handle lights positioned inside the model space and handle spotlights. Uniform grids are best forreflection and refraction rays with little coherence. We propose an Enforced Coherence method to bring coherence to them byrearranging the ray to voxel mapping using sorting. This gives the best performance on GPUs with only user-managed caches. Wealso propose a simple, Independent Voxel Walk method, which performs best by taking advantage of the L1 and L2 caches on recentGPUs. We achieve over 10 fps of total rendering on the Conference model with one light source and one reflection bounce, whilerebuilding the data structure for each stage. Ideas presented here are likely to give high performance on the future GPUs as well asother manycore architectures