Search

In much of the literature on the information society, its defining characteristic is assumed to be its immateriality. That is to say, as our interactions and activities become less dependent [...]

Author(s): Hirokazu Shirado [1, 2]; Nicholas A. Christakis (corresponding author) [1, 2, 3, 4] Coordination in groups faces a sub-optimization problem [1, 2, 3, 4, 5, 6] and theory suggests [...]

The 2010s have seen breakthroughs in frontiers from gene editing to gravitational waves. The coming one must focus on climate change. The 2010s have seen breakthroughs in frontiers from gene editing to gravitational waves. The coming one must focus on climate change., Author Affiliations: The CRISPR-Cas9 system that is used to cut DNA has changed biology forever. [see PDF for image] Credit: Juan Gaertner/SPL CRISPR-Cas9 gene editing complex The scientific events that [...]

Support vector machine (SVM) active learning is one popular and successful technique for relevance feedback in content-based image retrieval (CBIR). Despite the success, conventional SVM active learning has two main drawbacks. First, the performance of SVM is usually limited by the number of labeled examples. It often suffers a poor performance for the small-sized labeled examples, which is the case in relevance feedback. Second, conventional approaches do not take into account the redundancy among examples, and could select multiple examples that are similar (or even identical). In this work, we propose a novel scheme for explicitly addressing the drawbacks. It first learns a kernel function from a mixture of labeled and unlabeled data, and therefore alleviates the problem of small-sized training data. The kernel will then be used for a batch mode active learning method to identify the most informative and diverse examples via a min-max framework. Two novel algorithms are proposed to solve the related combinatorial optimization: the first approach approximates the problem into a quadratic program, and the second solves the combinatorial optimization approximately by a greedy algorithm that exploits the merits of submodular functions. Extensive experiments with image retrieval using both natural photo images and medical images show that the proposed algorithms are significantly more effective than the state-of-the-art approaches. A demo is available at http://msm.cais.ntu.edu.sg/LSCBIR/. Categories and Subject Descriptors: H.3.3 [Information Storage and Retrieval]: Information Search and Retrieval; H.4.0 [Information Systems Applications]: General; 1.2.6 [Artificial Intelligence] General Terms: Algorithms, Design, Experimentation, Performance Additional Key Words and Phrases: Content-based image retrieval, support vector machines, active learning, semisupervised learning, batch mode active learning, human-computer interaction ACM Reference Format: Hoi, S. C. H., Jin, R., Zhu, J., and Lyu, M. R. 2009. Semisupervised SVM batch mode active learning with applications to image retrieval. ACM Trans. Inform. Syst. 27, 3, Article 16 (May 2009), 29 pages. DOI = 10.1145/1508850.1508854 http://doi.acm.org/10.1145/1508850.1508854

The area of web usability has long intrigued researchers. It has been widely accepted that for a website to be successful, the level of usability has to be high. The reason is because of poorly designed website. While many web designers largely fall on their prior experience to tell if a web site design is good or bad, there is a need to be able to provide clear methodology to do so (2), (3). An article by Goldsborough which quoted Nielson, reported that while many web designers think their websites are important, generally it is users who will determine the relevance of these websites. In this paper, we discuss the methodology of usability and the survey conducted for the webpage that we have done. Key words: Web usability, consistency, interactivity, accessibility, content, INTRODUCTION HCI has developed to become an important area of computer science. Generally, HCI refers to a study of communication between human and computer which draws from supporting knowledge on [...]

This paper describes the inter-rater and intra-rater reliability, and the concurrent validity of an observational instrument, the Keyboard Personal Compute, Style instrument (K-PeCS), which assesses stereotypical postures and movements associated with computer keyboard use. Three trained raters independently rated the video clips of 45 computer keyboard users to ascertain inter-rater reliability, and then re-rated a sub-sample of 15 video clips to ascertain intra-rater reliability. Concurrent validity was assessed by comparing the ratings obtained using the K-PeCS to scores developed from a 3D motion analysis system. The overall K- PeCS had excellent reliability [inter-rater: intra-class correlation coefficients (ICC) = .90; intra-rater: ICC = .92]. Most individual items on the K-PeCS had from good to excellent reliability, although six items fell below ICC = .75. Those K-PeCS items that were assessed for concurrent validity compared favorably to the motion analysis data for all but two items. These results suggest that most items on the K-PeCS can be used to reliably document computer keyboarding style. Keywords: VDT evaluation; MSD-UE; Instrument development