Abstract In this paper we investigate the impact that altering the convolutional layer topology has upon the performance of computer vision tasks using a variety of widely used benchmark image datasets. Despite the widespread convention in convolutional neural networks, of incrementally doubling the filter count at each layer, there is little evidence substantiating the superiority of this method over other possible topologies. Our research reveals that a contrarian strategy—reducing the filters by half—can achieve performance on par with, if not superior to, this usual approach. We have extended our investigation to include a variety of novel topological structures. These empirical results challenge the prevailing assumption, that the sequential doubling of number of filters in the network configuration will always yield the best results with all datasets. Our findings advocate for a more nuanced approach to neural network design, incorporating a flexible approach to filter topologies into workflows. This could potentially have a significant impact upon the architectural standards in deep learning for visual recognition tasks.

It has been estimated that 85.5% ([14]) of school aged children own a mobile phone in the UK from which 29%([10]) of them access the Internet. It is assumed that this ownership provides the children with the opportunity to establish and maintain social connections with their fellow peers. According to Vygotsky's Social Development Theory([21]) this social interaction with peers plays a role in children's cognitive development that has been exploited in modern educational systems in various ways. This ongoing research aims to make further use of this association between ownership of mobile devices, social nteraction and learning by ultimately introducing a technological solution for children to communicate with their peers and access peer generated content. It will investigate whether the mobility and the information source, peer versus non-peer, enhances children's learning. Two separate investigations have been conducted, so far, to determine the sources and the tools children use to obtain support for their learning. A survey of 74 Key Stage 3 children and two focus group sessions with 34 similar age children found that they actively use technologies, such as Face book, SMS and email, to establish peer interaction.

With the expanding horizons of distributed eLearning, classrooms are increasingly consisting of a mix of different cultures. This paper addresses this factor in adaptive educational systems by presenting a new facet of the standard user model, the CAE-L cultural ontology framework. The definition and creation of the framework is introduced along with the first instances of its use and application as a subset of an AEH user model. This is further illustrated through the use of a CAE-L web service.

In this article we describe a hybrid technique for dictionary-based query translation suitable for English-Chinese cross language information retrieval. This technique marries a graph-based model for the resolution of candidate term ambiguity with a pattern-based method for the translation of out-of-vocabulary (OOV) terms. We evaluate the performance of this hybrid technique in an experiment using several NTCIR test collections. Experimental results indicate a substantial increase in retrieval effectiveness over various baseline systems incorporating machine- and dictionary-based translation.

This paper describes how visual and verbal learning styles have been successfully integrated into an adaptive educational environment. User trials of this system were carried out, to determine the effect of the adaptation, and although these user trials do not indicate any statistically significant differences, the qualitative information gleaned from the study indicated that students preferred using this environment over other traditional revision methods, and that they perceived personalized tuition to be better than a 'one size fits all' approach

The mechanisms for personalisation used in web applications are currently the subject of much debate amongst researchers from many diverse subject areas. One of the most contemporary ideas for user modelling in web applications is that of cognitive styles, where a user's psychological preferences are assessed stored in a database and then used to provide personalised content and/or links. We describe user trials of a case study that utilises visual-verbal preferences in an adaptive web-based educational system (AWBES). Students in this trial were assessed by the Felder-Solomon Inventory of Learning Styles (ILS) instrument, and their preferences were used as a means of content personalisation.Contrary to previous findings by other researchers, we found no significant differences in performance between matched and mismatched students. Conclusions are drawn about the value and validity of using cognitive styles as a way of modelling user preferences in educational web applications.