Narrow Results

A problem of automatic detection of tracks in aerial photos is considered. We adopt a Bayesian approach and base our inference on an a priori knowledge of the structure of tracks. The probability of a pixel to belong to a track depends on how the pixel gray level differs from the gray levels of pixels in the neighborhood and on additional prior information. Several suggestions on how to formalize the prior knowledge about the shape of the tracks are made. The Gibbs sampler is used to construct the most probable configuration of tracks in the area. The method is applied to aerial photos with cell size of 1 sq. m. Even for detection of trails of width comparable with or smaller than the cell size, positive results can be achieved.

We present a systematic and straightforward approach to the problem of single-trial classification of event-related potentials (ERP) in EEG. Instead of using a generic classifier off-the-shelf, like a neural network or support vector machine, our classifier design is guided by prior knowledge about the problem and statistical properties found in the data. In particular, we exploit the well-known fact that event-related drifts in EEG potentials, albeit hard to detect in a single trial, can well be observed if averaged over a sufficiently large number of trials. We propose to use the average signal and its variance as a generative model for each event class and use Bayes' decision rule for the classification of new and unlabeled data. The method is successfully applied to a data set from the NIPS*2001 Brain–Computer Interface post-workshop competition. Our result turned out to be competitive with the best result of the competition.

The application of machine learning methods in the engineering of intelligent technical systems often requires the integration of continuous constraints like positivity, monotonicity, or bounded curvature in the learned function to guarantee a reliable performance. We show that the extreme learning machine is particularly well suited for this task. Constraints involving arbitrary derivatives of the learned function are effectively implemented through quadratic optimization because the learned function is linear in its parameters, and derivatives can be derived analytically. We further provide a constructive approach to verify that discretely sampled constraints are generalized to continuous regions and show how local violations of the constraint can be rectified by iterative re-learning. We demonstrate the approach on a practical and challenging control problem from robotics, illustrating also how the proposed method enables learning from few data samples if additional prior knowledge about the problem is available.

This study draws from learning and opportunity development theories to develop and test a conceptual model of opportunity development. Using a sample of technology entrepreneurs in university incubators we explore the unique and joint effects of prior knowledge and learning on sales expectations and product innovativeness. Results suggest that domain knowledge of technology and demand are of unequal value to opportunity development. Domains of prior knowledge relate differently to outcomes but demand learning relates to sales expectations and product innovativeness. Findings suggest the types and sequence of knowledge are important to understanding opportunity development.

A structural model for the estimation of software failure rates is proposed which is based on a partition of the code into components. Systematic failure is assumed to be induced by the interaction of these components. A Bayesian inference scheme is used to perform failure rate estimation on the basis of N failure free test runs. The approach splits the problem of constructing a system prior up into the smaller, conceivably simpler problems of constructing subtask priors. Thereby a wider range of prior information is used in the process of assessing system safety. The work in this paper constitutes a first step towards a formal statistical understanding of the relationship between system complexity and system testing for reliability. Long-term implications are the achievement of more informative reliability estimates and guidelines on cost-effective testing.

In this paper, a modified least squares support vector machine classifier, called the C-variable least squares support vector machine (C-VLSSVM) classifier, is proposed for credit risk analysis. The main idea of the proposed classifier is based on the prior knowledge that different classes may have different importance for modeling and more weight should be given to classes having more importance. The C-VLSSVM classifier can be obtained by a simple modification of the regularization parameter, based on the least squares support vector machine (LSSVM) classifier, whereby more weight is given to errors in classification of important classes, than to errors in classification of unimportant classes, while keeping the regularized terms in their original form. For illustration purpose, two real-world credit data sets are used to verify the effectiveness of the C-VLSSVM classifier. Experimental results obtained reveal that the proposed C-VLSSVM classifier can produce promising classification results in credit risk analysis, relative to other classifiers listed in this study.

In general, it could be said that bats are blind but they have high quality senses of smelling and hearing to survive. Similarly, entrepreneurs can look at the business world with different eyes to survive. This affects their cognitive biases in risk perception. The aim of this study was to analyze how entrepreneurs’ cognitive biases affect their opportunity exploitation and risk perception. In this study, we evaluated self-efficacy, locus of control, overconfidence and optimism as dimensions of cognition. Independently of our purpose, results also show entrepreneurs have social capital, such as experience and prior knowledge, which forms their cognitive biases and leads them to perceive less risk when evaluating a new venture idea.

Image segmentation is often required as a fundamental stage in medical image processing, particularly during the clinical analysis of magnetic resonance (MR) brain images. Fuzzy c-means (FCM) clustering algorithm is one of the best known and widely used segmentation methods, but this algorithm has some problem for segmenting simulated MRI images to high number of clusters with different noise levels and real images because of spatial complexities. Anatomical segmentation usually requires information derived from the manual segmentations done by experts, prior knowledge can be useful to modify image segmentation methods. In this paper, we propose some methods to modify FCM algorithm using expert manual segmentation as prior knowledge. We developed combination of FCM algorithm and prior knowledge in three ways, in order to improve segmentation of brain MR images with high noise level and spatial complexity. In real images, we had a considerable improvement in similarity index of three classes (white matter, gray matter, CSF) and in simulated images with different noise levels evaluation criteria of white matter and gray matter has improved.

This chapter presents practical ways to use the whiteboard more effectively to foster mathematics learning. The constructivist’s view of learning advocates that knowledge must be built upon prior knowledge for learning to be relevant and meaningful. To foster mathematics learning, the whiteboard can be exploited as an instructional scaffold for sequencing mathematics visually so that students will have greater clarity of where the lesson starts, the connections within the lesson, and where the lesson is going. This coherent progression of a lesson facilitates students to organize their thinking and help students see the connections between different parts of the lesson. The use of the whiteboard to show connections between various mathematics concepts, present contrasting solutions and as a record of key concepts will be illustrated in this chapter.

The teaching through problem solving approach uses problem solving as a means to stimulate students’ learning of mathematics. Students work either individually or in groups on an open problem before the concept is taught to the class. Teachers act as facilitators, asking probing questions to help students reflect on their thinking. This approach of teaching allows students to explore and think about a problem with a realistic context and discover knowledge in the process and is in contrast to a didactic teaching approach, where the teacher explicitly tells students what they are to learn. The opportunity to explore and think provided by the teaching through problem solving approach may better prepare students to thrive in a rapidly changing world, as opposed to students receiving information directly from teachers. This chapter discusses the teaching through problem solving approach, its benefits and its challenges. A lesson carried out by the author using this approach will be discussed.

Spoken Language Understanding (SLU), which is a crucial part of spoken dialogue systems, includes slot filling (SF) and intent detection (ID) tasks which are a high degree of correlation and influence each other. A joint learning model can share information between SF and ID to effectively improve experimental results. However, most exciting models do not use the information between SF and ID, making the model perform poorly. In this paper, we propose a novel based on the prior knowledge joint learning model for better utilizing the semantic information between SF and ID. The experimental results on three public datasets show that based on prior knowledge joint learning model can better express sentence semantic information and improve the accuracy of the ID and SF tasks.

This study explores the mythical qualities of learning mathematics in China by addressing the unique characteristics of Chinese mathematics teaching and learning: using the learning-questioning and learning-reviewing instructional model to enhance students' understanding of mathematics concepts and reinforce mathematics proficiency, Furthermore, this study investigates Chinese teachers' beliefs in using the learning-questioning and learning-reviewing instructional model and their impact on teaching and learning mathematics. Four fifth-grade mathematics teachers from four schools in a large city in Jiangsu Province participated in this study. Data were collected via classroom observations and interviews with participating teachers. Findings indicate that Chinese teachers were able to apply various methods to help students learn mathematics, focusing on questioning and reviewing strategies and helping students to understand mathematics concepts and develop mathematics proficiency. Furthermore, Chinese teachers' beliefs have a deep impact on their teaching practices: teachers believe that conceptual understanding and procedural development are equally important in mathematics teaching; teaching mathematics is teaching the thinking method.