An online transfer learning RBF neural network for cross domain data classification

In this paper, a Radial Basis Function Network (RBFN) trained with the Dynamic Decay Adjustment (DDA) algorithm (i.e., RBFNDDA) is deployed as an incremental learning model for tackling transfer learning problems. An online learning strategy is exploited to allow the RBFNDDA model to transfer knowledge from one domain and applied to classification tasks in a different yet related domain. An experimental study is carried out to evaluate the effectiveness of the online RBFNDDA model using a benchmark data set obtained from a public domain. The results are analyzed and compared with those from other methods. The outcomes positively reveal the potentials of the online RBFNDDA model in handling transfer learning tasks. © 2014 The authors and IOS Press. All rights reserved.

Distributed Q-learning controller for a multi-intersection traffic network

This paper proposes a Q-learning based controller for a network of multi intersections. According to the increasing amount of traffic congestion in modern cities, using an efficient control system is demanding. The proposed controller designed to adjust the green time for traffic signals by the aim of reducing the vehicles’ travel delay time in a multi-intersection network. The designed system is a distributed traffic timing control model, applies individual controller for each intersection. Each controller adjusts its own intersection’s congestion while attempt to reduce the travel delay time in whole traffic network. The results of experiments indicate the satisfied efficiency of the developed distributed Q-learning controller.

Rock slope stability analyses using extreme learning neural network and terminal steepest descent algorithm

The analysis of rock slope stability is a classical problem for geotechnical engineers. However, for practicing engineers, proper software is not usually user friendly, and additional resources capable of providing information useful for decision-making are required. This study developed a convenient tool that can provide a prompt assessment of rock slope stability. A nonlinear input-output mapping of the rock slope system was constructed using a neural network trained by an extreme learning algorithm. The training data was obtained by using finite element upper and lower bound limit analysis methods. The newly developed techniques in this study can either estimate the factor of safety for a rock slope or obtain the implicit parameters through back analyses. Back analysis parameter identification was performed using a terminal steepest descent algorithm based on the finite-time stability theory. This algorithm not only guarantees finite-time error convergence but also achieves exact zero convergence, unlike the conventional steepest descent algorithm in which the training error never reaches zero.

Visual feedback control of a robot in an unknown environment (learning control using neural networks)

In this paper, a visual feedback control approach based on neural networks is presented for a robot with a camera installed on its end-effector to trace an object in an unknown environment. First, the one-to-one mapping relations between the image feature domain of the object to the joint angle domain of the robot are derived. Second, a method is proposed to generate a desired trajectory of the robot by measuring the image feature parameters of the object. Third, a multilayer neural network is used for off-line learning of the mapping relations so as to produce on-line the reference inputs for the robot. Fourth, a learning controller based on a multilayer neural network is designed for realizing the visual feedback control of the robot. Last, the effectiveness of the present approach is verified by tracing a curved line using a 6-degrees-of-freedom robot with a CCD camera installed on its end-effector. The present approach does not necessitate the tedious calibration of the CCD camera and the complicated coordinate transformations.

Young children learning about evaporation: a longitudinal perspective

This paper traces the learning pathways over a 4-year period of 12 children learning about evaporation. The findings show the complexity and dependence on context of children's understandings. Detailed transcripts for 2 children are used to demonstrate how understandings of phenomena are framed within a network of personal narratives of self that reflect children's different subjectivities as learners and school children, It is argued that the longitudinal methodology opens up a more complex and nuanced view of children's conceptual learning in school settings than is afforded by cross-sectional studies and that the focus on individuals over time compels a very different construction of the learner than is represented in mainstream conceptual-change literature. [ABSTRACT FROM AUTHOR]

Learning and teaching assessment: reviewing the evidence

The authors have recently completed a research review on learning and teaching of assessment in social work which was commissioned by the Social Care Institute for Excellence (SCIE) and the Social Policy and Social Work Learning and Teaching Support Network (SWAPltsn) to support the development of the new social work award in England. This involved reviewing relevant literature from social work and cognate disciplines back to 1990 with the aim of identifying best practice in learning and teaching of assessment skills.

Although assessment has been recognised as a core skill in social work and should underpin social work interventions, there is no singular theory or understanding as to what the purpose of assessment is and what the process should entail. Social work involvement in the assessment process may include establishing need or eligibility for services, to seek evidence of past events or to determine likelihood of future danger, may underpin recommendations to other agencies, or may determine the suitability of other service providers. In some settings assessment is considered to begin from the first point of contact and may be a relatively short process, whereas elsewhere it may be a process involving several client contacts over an extended period of time. The assessment process may range from the collection of data on standardised proforma to a flexible approach depending on circumstances. These variations permeate the literature on the learning and teaching of assessment in social work and cognate disciplines.

Several different approaches to classroom based learning were proposed in the literature including case-based teaching, interviews with actors who have been trained to play 'standardised clients', and observation of children and families, as well as didactic lecturing and various uses of video equipment and computers. Furthermore learning by doing has long been one of the hallmarks of social work education, and there are a number of models proposed in which students learn about the assessment process through conducting assessments. The evidence to support these different approaches to learning and teaching is variable. Based on the evidence reviewed, recommendations as to what is good practice in learning and teaching about assessment will be presented.

The online learning alchemist : preventing gold turning into lead

This paper addresses the creation of materials and resources for use in online learning, focusing on the new and emerging roles for teachers and learners in conjunction with developments in our understanding of the human-computer interface. As more educational providers adopt network-based technologies as delivery portals, the demand for skills in the creation of effective online resources is becoming critical. If we are to provide the learner with online resources that will enhance knowledge construction and the teacher with clear measures that these activities are effective, then we as resource developers must resurrect the role of what might be termed the online alchemist. Our first task is to ensure that new digital resources are not simply transferred from their original format but repurposed to ensure learner(s) accessing those resources are able to interact with both the content and their collaborative partners with new levels of flexibility and manipulation. We must transcend the too frequent use of technology as a means to replicate existing resources and conceptualise environments that engender new paradigms for teaching and learning. Our challenge remains to ensure the gold we have in effective teaching strategies and learning resources is not tarnished through ineffective applications within the online learning context. One strategy to achieve this is through proactive evaluation, a framework that integrates a set of factors and influences to better inform the development of online learning resources.

An immune network approach for web document clustering

The human immune system provides inspiration for solving a wide range of innovative problems. In this paper, we propse an immune network based approach for web document clustering. All the immune cells in the network competitively recognize the antigens (web documents) which are presented to the network one by one. The interaction between immune cells and an antigen leads to an augment of the network through the clonal selection and somatic mutation of the stimulated immune cells, while the interaction among immune cells results in a network compression. The structure of the immune network is well maintained by learning and self-regularity. We use a public web document data set to test the effectiveness of our method and compare it with other approaches. The experimental results demonstrate that the most striking advantage of immune-based data clustering is its adaptation in dynamic environment and the capability of finding new clusters automatically.

Adopting web-based learning and teaching: a case study in higher education

Most universities worldwide are becoming distance education providers through adopting web-based learning and teaching via the introduction of learning management systems that enable them to open their courses to both on- and off-campus students. Whether this is an effective introduction depends on factors that enable and impede the adoption of such systems and their related pedagogical strategies. This study examines such factors related to adopting a learning management system in a large multicampus urban Australian university. The research method used case study approaches and purposively selected the sample consisting of innovative teaching academics from across the university, who used web-based approaches to teach both on- and off-campus learners. The data were analyzed using a combination of Rogers' theory of diffusion of innovations and actor-network theory and revealed a series of enabling and impeding factors faced by pioneering technology-adopter teaching academics, some of which are technology related while others are policy related and common to large multicampus institutions. The study found that safe adoption environments recognizing career priorities of academics are a result of the continuous negotiation between the evolving institution and its innovative and creative staff. The article concludes with a series of conditions that would form a safe, enabling, and encouraging environment for technology-adopter teaching academics in a large multicampus higher education setting.

Cultural factors behind the adoption of e-learning in Turkey

This paper aims to examine ways in which cultural factors shape the adoption and use of information technology for online teaching. This research focuses on influential early adopters in the tertiary education sector in Turkey who have become change-agents by inspiring small networks of their peers. The study examines the operation of trust and inspiration in networking and teamwork in the Asian academic environment. Findings from this research can assist individuals and institutions to better understand ways in which to optimize the online teaching and learning experience for staff.

The application of rough set and Kohonen network to feature selection for object extraction

Selecting a set of features which is optimal for a given task is a problem which plays an important role in a wide variety of contexts including pattern recognition, images understanding and machine learning. The paper describes an application of rough sets method to feature selection and reduction in texture images recognition. The proposed methods include continuous data discretization based on Kohonen neural network and maximum covariance, and rough set algorithms for feature selection and reduction. The experiments on trees extraction from aerial images show that the methods presented in this paper are practical and effective.

LAN cafes: cafes, places of gathering or sites of informal teaching and learning?

Despite the interest of sociologists and educational researchers in Internet café as sites for new cultural and social formations and informal learning, thus far little attention has been paid to the function of café owners, managers and other staff in the mediation and co-construction of those spaces. Drawing from interviews with managers of commercial Internet café in Australia specialising in LAN (Local Area Network) gaming, this article seeks to examine their role and their attitudes more closely; in particular with regard to school-aged users of their facilities. We contend that LAN café are liminal spaces situated at the margins of Australian culture and located at the junctions between home, school and the street, online and offline spaces, work and play. The roles of LAN café managers are similarly ambiguous: in many ways they can be regarded as informal teachers facilitating the process of informal learning.

Reactions and responses : the concept of learning networks

In Victoria, Australia, the potential of networks to respond to issues of risk for young people has been identified. Learning networks have become a means to reform resulting in a number of policy strategies by the Victorian Labour government since 2000 and including School Networks and Local Learning and Employment Networks (LLENs). At the same time organic learning networks continue to form at a ‘grass roots’ level as communities respond to perceived needs. This presentation draws on research into the foundation, formation and practice of the Smart Geelong Region Local Learning and Employment Network (SGR LLEN) and considers the opportunities and tensions when government policies take up and institutionalise solutions in ways that potentially work against the conditions that make them effective, viable and sustainable.

A stable MCA learning algorithm

Minor component analysis (MCA) is an important statistical tool for signal processing and data analysis. Neural networks can be used to extract online minor component from input data. Compared with traditional algebraic  approaches, a neural network method has a lower computational complexity. Stability of neural networks learning algorithms is crucial to practical applications. In this paper, we propose a stable MCA neural networks learning algorithm, which has a more satisfactory numerical stability than some existing MCA algorithms. Dynamical behaviors of the proposed algorithm are analyzed via deterministic discrete time (DDT) method and the conditions are obtained to guarantee convergence. Simulations are carried out to illustrate the theoretical results achieved.

Cultural factors affecting the adoption of e-learning in Turkey

This paper reports findings from an ethnographic study of e-learning adopters in Turkey and examines ways in which cultural factors shape the adoption and use of information technology for online teaching. This research focuses on influential early adopters in the tertiary education sector in Turkey who have become change-agents by inspiring small networks of their peers. The study examines the operation of trust and inspiration in networking and teamwork in the Asian academic environment. The key findings of this research are that the early adopters become change agents in small groups and networks and that the process of adoption relies heavily on social networks and connections. Findings from this research can assist individuals and institutions to better understand ways in which to optimize the online teaching and learning experience for staff.