Yahoo!

About: Yahoo! is a company organization based out in London, United Kingdom. It is known for research contribution in the topics: Population & Web search query. The organization has 26749 authors who have published 29915 publications receiving 732583 citations. The organization is also known as: Yahoo! Inc. & Maudwen-Yahoo! Inc.

Learning from multiple annotators with varying expertise

Abstract: Learning from multiple annotators or knowledge sources has become an important problem in machine learning and data mining. This is in part due to the ease with which data can now be shared/collected among entities sharing a common goal, task, or data source; and additionally the need to aggregate and make inferences about the collected information. This paper focuses on the development of probabilistic approaches for statistical learning in this setting. It specially considers the case when annotators may be unreliable, but also when their expertise vary depending on the data they observe. That is, annotators may have better knowledge about different parts of the input space and therefore be inconsistently accurate across the task domain. The models developed address both the supervised and the semi-supervised settings and produce classification and annotator models that allow us to provide estimates of the true labels and annotator expertise when no ground-truth is available. In addition, we provide an analysis of the proposed models, tasks, and related practical problems under various scenarios. In particular, we address how to evaluate annotators and how to consider cases where some ground-truth may be available. We show experimentally that annotator expertise can indeed vary in real tasks and that the presented approaches provide clear advantages over previously introduced multi-annotator methods, which only consider input-independent annotator characteristics, and over alternative approaches that do not model multiple annotators.

Glenohumeral chondrolysis after shoulder arthroscopy: case reports and review of the literature

Abstract: A 16-year-old male baseball and football player complained of pain in his right, dominant shoulder after 3 episodes of subluxation over the course of 2 years secondary to diving on the football and baseball fields. He was treated conservatively after each but failed to improve with 4 months of rehabilitation after the third subluxation. He underwent right shoulder arthroscopy in April 2001 at an outside institution for debridement of a posteroinferior labral tear. Figure 1 shows a preoperative radiograph of the affected shoulder. A radiofrequency ablator was used for this debridement. An extensive synovectomy and a coracoacromial ligament resection were also performed with the ablator followed by a modified subacromial decompression. His postoperative course was complicated by poor return of glenohumeral motion and mild to moderate pain with activities of daily living (ADLs). He was unable to return to sports. His glenohumeral joint was injected 4 months after the procedure with steroid and marcaine. He reported good immediate relief of pain following the injection with relief lasting for only 24 hours. Plain radiography was repeated at 5 months postoperative (Figure 2). It demonstrated significant glenohumeral space narrowing with early subchondral cyst formation in the glenoid. He presented to our clinic for the first time the following week. Range of motion (ROM) at that time was forward elevation to 180°, external rotation with elbow at the side to 60° and internal rotation to L4, and 1+ anterior laxity. Physical therapy was continued, focusing on maintaining glenohumeral ROM. A rheumatologic evaluation was also recommended. It was subsequently normal. He returned again to our clinic 8 months after his shoulder arthroscopy. He described a slow progression of pain in his right shoulder since his last visit, now with substantial pain during ADLs. Plain radiographs demonstrated progression of the glenohumeral joint space narrowing (Figure 3). An MRI demonstrated thinning of the glenohumeral articular surfaces and subchondral cyst formation in the glenoid and humerus. A repeat arthroscopy was performed as a diagnostic maneuver to rule out infection and to obtain a synovial biopsy. Arthroscopic findings were as follows: the glenoid was devoid of articular cartilage, and the exposed subchondral bone appeared friable (Figure 4). The humeral head was covered with smooth hyaline cartilage, but marginal osteophytes were present. The synovium was red with numerous villous projections throughout. The labrum was degenerative circumferentially, also with villous fibrillations throughout (Figure 5). Operative cultures were negative for aerobes, anaerobes, acid-fast bacilli, and fungus. Biopsy of the synovium demonstrated numerous villous projections with polygonal to round stromal cells, some of which contained hemosiderin pigment. Rare multinucleated giant cells were present, with few chronic inflammatory cells and early pannus formation (Figure 6). Glenohumeral Chondrolysis After Shoulder Arthroscopy

Learning Graph Matching

Abstract: As a fundamental problem in pattern recognition, graph matching has applications in a variety of fields, from computer vision to computational biology. In graph matching, patterns are modeled as graphs and pattern recognition amounts to finding a correspondence between the nodes of different graphs. Many formulations of this problem can be cast in general as a quadratic assignment problem, where a linear term in the objective function encodes node compatibility and a quadratic term encodes edge compatibility. The main research focus in this theme is about designing efficient algorithms for approximately solving the quadratic assignment problem, since it is NP-hard. In this paper we turn our attention to a different question: how to estimate compatibility functions such that the solution of the resulting graph matching problem best matches the expected solution that a human would manually provide. We present a method for learning graph matching: the training examples are pairs of graphs and the `labels' are matches between them. Our experimental results reveal that learning can substantially improve the performance of standard graph matching algorithms. In particular, we find that simple linear assignment with such a learning scheme outperforms Graduated Assignment with bistochastic normalisation, a state-of-the-art quadratic assignment relaxation algorithm.

Rain Attenuation in Millimeter Wave Ranges

Abstract: The high potential of millimeter-wave communication systems has generated the need to carry out many studies in view of rain and other climatic effects on radio propagation at these frequencies. This paper reviews rain attenuation in millimeter wave ranges. In the present study, a short-range 35 GHz radio link was used to measure rain specific attenuation with simultaneous measurement of rain rate distribution. The rainfall statistics and attenuation caused by rains are discussed, and an empirical model derived from these measurements is suggested in order to observe and investigate the attenuation caused by rains in short-range communications. A millimeter wave propagation experiment at 103 GHZ on a propagation path of 390 m is conducted. The results were compared with the rain attenuation calculations from the Marshall-Palmer, Best, Joss-Thomas-Waldvogel and Weibull distributions for raindrop size. It has been shown that the Weibull distribution has a good agreement with the experiments. Finally the analysis and discussion for measurement results respectively.

International English: A guide to varieties of Standard English

Abstract: Peter Trudgill & Jean Hannah, International English: A guide to varieties of Standard English. 4th ed. London: Arnold; New York: Oxford University Press, 2002. Pp. i–xv, 1–153. Pb $22.95.This book provides a delightful survey of the global variety of pronunciation and usage of English as an educated standard. It focuses on the phonetics of the various Englishes, especially on the vowels, where so much of the variability resides, and on differences in usage, lexical and syntactic as well as orthographic. Although this small volume necessarily deals with most topics briefly, it includes a wealth of detail.