Task Setup Research Articles

Batch Mode Active Learning with Applications to Text Categorization and Image Retrieval

Most machine learning tasks in data classification and information retrieval require manually labeled data examples in the training stage. The goal of active learning is to select the most informative examples for manual labeling in these learning tasks. Most of the previous studies in active learning have focused on selecting a single unlabeled example in each iteration. This could be inefficient, since the classification model has to be retrained for every acquired labeled example. It is also inappropriate for the setup of information retrieval tasks where the user's relevance feedback is often provided for the top K retrieved items. In this paper, we present a framework for batch mode active learning, which selects a number of informative examples for manual labeling in each iteration. The key feature of batch mode active learning is to reduce the redundancy among the selected examples such that each example provides unique information for model updating. To this end, we employ the Fisher information matrix as the measurement of model uncertainty, and choose the set of unlabeled examples that can efficiently reduce the Fisher information of the classification model. We apply our batch mode active learning framework to both text categorization and image retrieval. Promising results show that our algorithms are significantly more effective than the active learning approaches that select unlabeled examples based only on their informativeness for the classification model.

Automatic medical image annotation in ImageCLEF 2007: Overview, results, and discussion

In this paper, the automatic medical annotation task of the 2007 CLEF cross language image retrieval campaign (ImageCLEF) is described. The paper focusses on the images used, the task setup, and the results obtained in the evaluation campaign. Since 2005, the medical automatic image annotation task exists in ImageCLEF with increasing complexity to evaluate the performance of state-of-the-art methods for completely automatic annotation of medical images based on visual properties. The paper also describes the evolution of the task from its origin in 2005–2007. The 2007 task, comprising 11,000 fully annotated training images and 1000 test images to be annotated, is a realistic task with a large number of possible classes at different levels of detail. Detailed analysis of the methods across participating groups is presented with respect to the (i) image representation, (ii) classification method, and (iii) use of the class hierarchy. The results show that methods which build on local image descriptors and discriminative models are able to provide good predictions of the image classes, mostly by using techniques that were originally developed in the machine learning and computer vision domain for object recognition in non-medical images.

Mathematics-related tasks, teacher education, and teacher educators

ly about mathematics used to solve problems and were able to use successfully the techniques learned (Bloom; Hsu et al.); their ability to identify structure by analogical thinking increased (Peled; Yackel et al.); their production of correct or even improved solutions and analyses developed (Liljedahl et al.; Karp); with respect to curricular coherence, Ferrini-Mundy et al. found that ‘‘the task led them [the teachers] into discussions of mathematical progressions of ideas, increasingly more general ways of using fractions to reason about situations, and connections among symbolic and other representations’’. Silver et al. consider teachers’ engagements in their tasks as ‘‘significant opportunities to learn mathematics and to build or strengthen connections among related mathematical ideas’’. Teachers also gained better perceptions of, and insights into, various aspects related to their practice, e.g., what is required to prevent and ameliorate misconceptions and how to confront the difficulties in a mathematical task (Baturo et al.); unexpected difficulties revealed by the comparison between the set-up of tasks and their implementation (Horoks and Robert); they organized a complex constellation of mathematical ideas, making connections to students’ thinking about these ideas as well as to ‘‘a range of pedagogical actions and decisions that affect students’ opportunities to learn’’ (Silver et al.). Thompson et al. noticed that ‘‘teachers who became self-aware of the reasoning behind their successful predictions and explanations were then positioned to project that reasoning and its development onto images of their students’ thinking.’’ On the meta-level and dispositions, teachers increased their sensitivity to and awareness of several aspects of teaching mathematics, such as the connections through the curriculum as well as how choice of classroom tasks influence students’ experiences of the curriculum (Prestage and Perks) the value of systematic variation in exploratory tasks (Karp), the relationship between a task and the kind of thinking that the task may evoke in students (Ben-Chaim et al.). Teachers appeared to learn to reflect on their teaching and knowledge more often than they would have done otherwise (Peng). In short, there are strong claims regarding teacher learning. These claims are supported mainly by MTEs’ spontaneous or systematic observations and impressions. Several authors base their claims on other sources, such as MTs’ self-reports, written and oral performance on assignments, pre-post testing (Ben-Chaim et al.), MTs’ lessons—planned, taught, and 438 O. Zaslavsky

The CLEF 2005 Automatic Medical Image Annotation Task

In this paper, the automatic annotation task of the 2005 CLEF cross-language image retrieval campaign (ImageCLEF) is described. This paper focuses on the database used, the task setup, and the plans for further medical image annotation tasks in the context of ImageCLEF. Furthermore, a short summary of the results of 2005 is given. The automatic annotation task was added to ImageCLEF in 2005 and provides the first international evaluation of state-of-the-art methods for completely automatic annotation of medical images based on visual properties. The aim of this task is to explore and promote the use of automatic annotation techniques to allow for extracting semantic information from little-annotated medical images. A database of 10.000 images was established and annotated by experienced physicians resulting in 57 classes, each with at least 10 images. Detailed analysis is done regarding the (i) image representation, (ii) classification method, and (iii) learning method. Based on the strong participation of the 2005 campain, future benchmarks are planned.