Imbalanced classification in sparse and large behaviour datasets

Recent years have witnessed a growing number of publications dealing with the imbalanced learning issue. While a plethora of techniques have been investigated on traditional low-dimensional data, little is known on the effect thereof on behaviour data. This kind of data reflects fine-grained behaviours of individuals or organisations and is characterized by sparseness and very large dimensions. In this article, we investigate the effects of several over-and undersampling, cost-sensitive learning and boosting techniques on the problem of learning from imbalanced behaviour data. Oversampling techniques show a good overall performance and do not seem to suffer from overfitting as traditional studies report. A variety of undersampling approaches are investigated as well and show the performance degrading effect of instances showing odd behaviour. Furthermore, the boosting process indicates that the regularization parameter in the SVM formulation acts as a weakness indicator and that a combination of weak learners can often achieve better generalization than a single strong learner. Finally, the EasyEnsemble technique is presented as the method outperforming all others. By randomly sampling several balanced subsets, feeding them to a boosting process and subsequently combining their hypotheses, a classifier is obtained that achieves noise/outlier reduction effects and simultaneously explores the majority class space efficiently. Furthermore, the method is very fast since it is parallelizable and each subset is only twice as large as the minority class size.     

Expert system for problem solving in distance university education: The successful case of the subject “operations management”

Acquiring the skills needed to solve certain types of problems is one of the main challenges of distance university education, demanding extra motivation from students to tackle it. New technology should be one of our great allies in addressing these problems. This article proposes an expert system with a web‐based problem‐solving simulator for a multidevice environment in order to allow students to access an unlimited number of problems with their corresponding solutions, immediately, anytime and anywhere. The types of problems that can be used are those based on quantitative methods with a fixed methodology to be followed. A successful case study was carried out for the subject Operations Management at the Distance University of Madrid (UDIMA), with three different problem simulators having been developed. The results were very satisfactory compared with previous academic years in terms of motivation. A decrease in students that did not sit the final examination was observed, as was an increase in the number of visits to the Moodle classroom. Students also responded positively in their assessments of the initiative.     

盘中的意式美学

正现任金陵饭店太平洋西餐厅主厨的Nino,是一位意大利人,来金陵饭店工作快2年,用他的话说,这里是我的家。Nino非常喜欢中餐,做得一手美味的意大利菜,他总是有很多创意和想法,对食材的新鲜度要求严格,并注重摆盘的造型与色彩搭配,在心血来潮的时候会做出几道新菜,然后豪气地告诉小编尽情品尝,但是没有菜名!这可难住了每道菜必问菜名的小编,下面来欣赏一下没有菜名的意式美食,说不定能帮Nino起一个同样有美感的菜名。     

Compositional strategies with generative scores

Generative processes have found broad application in the design of innovative musical forms but it is only comparatively recently through the affordances of modern digital technologies that composers have been able to explore integrating such processes in the real-time creation of musical scores. This paper explores some of these affordances as well as a variety of approaches to some of the key aesthetic and compositional challenges through the analysis of recent works by composers Georg Hajdu, Ryan Ross Smith, Jason Freeman, David Kim-Boyle, and Arne Eigenfeldt.     

High‐speed mobility on planetary surfaces: A technical review

Despite the success so far accomplished in the robotic exploration of the Moon and Mars, the constraints associated with newly proposed mission concepts manifest the need for a faster surface prospection. Increasing driving velocities is being considered as a potential solution to the requirements introduced by these missions. This review presents the benefits and foreseeable challenges of using faster locomotive solutions for space exploration. Information is provided regarding the set of missions that would benefit most from faster locomotive capabilities. Starting by understanding the theoretical framework governing the interaction of wheeled robots operating over loose, sandy terrains, we delve into the foundation of Bekker's classic terramechanic equations—the most frequently used method to predict mobile robots off‐road performance. We highlight its limitations and review the efforts that have been made to expand the range of application of these theories to dynamic wheel–soil interactions. We analyze the existing experimental evidence on the effects of increasing traveling velocities under earthbound, off‐road conditions. By paying special attention to previous experiences on the lunar surface, we outline the challenges that the combination of irregular terrains and a reduced‐gravity field may pose to a fast‐moving exploration rover. The principles, mathematical models, experimental evidence, and experiences presented in this review are meant to aid in the identification of poorly understood and insufficiently studied aspects regarding high‐speed extraterrestrial surface mobility.     

Interactive and collaborative technological ecosystems for improving academic motivation and engagement

Universal Access in the Information Society -     

Sensitivity Analysis and Model Validation of a 2-DoF Mini Spherical Robot

This paper is focused on the development and validation of an error kinematic model of a mini spherical robot, aimed at its kinematic calibration. The robot is actually a spatial five-bar linkage, provided with two rotational degrees of freedom. A non-overconstrained kinematics is assumed for the robot in order to provide a simple mathematical model and allow a sensitivity analysis of all the involved geometric parameters. A simplified version of the model is proposed. It differs only for the degree of approximation. A comparison between full and reduced models is presented by means of numerical simulations, analyzing their behavior in a significant region of the robot workspace. In order to validate both of them a robot calibration is carried out on a physical prototype of the robot using a vision system, namely a fixed camera in a eye-to-hand configuration. An iterative algorithm aimed at the experimental identification of the geometric data of the robot is used. Some experimental results show the effectiveness of the study.     

Text and non-text separation in offline document images: a survey

International Journal on Document Analysis and Recognition (IJDAR) - Separation of text and non-text is an essential processing step for any document analysis system. Therefore, it is important to...     

Runtime verification of autopilot systems using a fragment of MTL- $${int }$$

International Journal on Software Tools for Technology Transfer - Current real-time embedded systems development frameworks lack support for the verification of properties using explicit time where...