Rebecca Strachan

Hybrid decision tree and naïve Bayes classifiers for multi-class classification tasks

In this paper, we introduce two independent hybrid mining algorithms to improve the classification accuracy rates of decision tree (DT) and naive Bayes (NB) classifiers for the classification of multi-class problems. Both DT and NB classifiers are useful, efficient and commonly used for solving classification problems in data mining. Since the presence of noisy contradictory instances in the training set may cause the generated decision tree suffers from overfitting and its accuracy may decrease, in our first proposed hybrid DT algorithm, we employ a naive Bayes (NB) classifier to remove the noisy troublesome instances from the training set before the DT induction. Moreover, it is extremely computationally expensive for a NB classifier to compute class conditional independence for a dataset with high dimensional attributes. Thus, in the second proposed hybrid NB classifier, we employ a DT induction to select a comparatively more important subset of attributes for the production of naive assumption of class conditional independence. We tested the performances of the two proposed hybrid algorithms against those of the existing DT and NB classifiers respectively using the classification accuracy, precision, sensitivity-specificity analysis, and 10-fold cross validation on 10 real benchmark datasets from UCI (University of California, Irvine) machine learning repository. The experimental results indicate that the proposed methods have produced impressive results in the classification of real life challenging multi-class problems. They are also able to automatically extract the most valuable training datasets and identify the most effective attributes for the description of instances from noisy complex training databases with large dimensions of attributes.

An adaptive ensemble classifier for mining concept drifting data streams

It is challenging to use traditional data mining techniques to deal with real-time data stream classifications. Existing mining classifiers need to be updated frequently to adapt to the changes in data streams. To address this issue, in this paper we propose an adaptive ensemble approach for classification and novel class detection in concept drifting data streams. The proposed approach uses traditional mining classifiers and updates the ensemble model automatically so that it represents the most recent concepts in data streams. For novel class detection we consider the idea that data points belonging to the same class should be closer to each other and should be far apart from the data points belonging to other classes. If a data point is well separated from the existing data clusters, it is identified as a novel class instance. We tested the performance of this proposed stream classification model against that of existing mining algorithms using real benchmark datasets from UCI (University of California, Irvine) machine learning repository. The experimental results prove that our approach shows great flexibility and robustness in novel class detection in concept drifting and outperforms traditional classification models in challenging real-life data stream applications.

Design of educational systems for work based learning (WBL): the learner experience

Purpose – Information and communication technologies have transformed higher education providing e‐mechanisms to support the delivery of content, communication and interaction. One example is the widespread adoption of virtual learning environments (VLEs) by higher education institutions to provide a key interface among learners, the content and tutors. The aim of this paper is to explore the effectiveness of the design of educational systems for work based learning (WBL) from the learners perspective. The study includes consideration of the use of technology to support the work based learning process for the learner.Design/methodology/approach – Case studies from four postgraduate programmes and one undergraduate programme within the School of Computing, Engineering and Information Sciences (CEIS) at Northumbria University are presented. Key results from a survey of students on these programmes are discussed.Findings – The results demonstrate that there are different motivations for this type of provisi...

Learning mathematics through serious games: An engagement framework

Digital games have become part of childhood and adolescence. The debate has moved from should teenagers play digital games to how to gain benefits from this gameplay. Researchers predict that technology-enhanced learning will increase with educational computer games (serious games) playing an important role. Although serious games are often built on established educational theories, they can also appear boring and struggle to engage the learner. Analyses of serious games demonstrate that many do not offer an entertainment experience comparable to or even recognizable as relatives of the entertainment games familiar to many players. However, a high level of engagement by the learner is viewed as necessary to provide a strong learning environment. The long-term aim of this study is to explore how digital games can support a more engaging and effective mathematics learning experience. This first stage has taken a qualitative grounded theory approach to explore the engagement factors of digital games among young people (aged 7 - 16 years). Analyses of data gathered through a combination of surveys and interviews have led to a framework of engagement factors. This provides a basis for designing serious games that are effective by being both engaging and educational.

Active Student Engagement: The Heart of Effective Learning

The chapter’s focal argument is that the format, duration and delivery of a course are most effective when the underlying philosophy is to engage students actively in their learning. This philosophy should be at the heart of the course’s design, development and delivery. The authors draw on their experiences of higher education in both the United Kingdom and Sri Lanka and across full-time, part-time, distance learning and work-based learning to illustrate that a variety of formats and delivery approaches are possible, but the crucial element is to ensure that students are active not passive learners. This theory aligns with a learner-centred, constructivist approach and lends itself to more authentic learning. Using examples from a number of different disciplines, the authors discuss the variation that can occur in course delivery and format whilst still encouraging and supporting an active learning approach. The final section of the chapter will focus on how this approach may require staff to adopt new methods of learning, teaching and assessment and their professional development plays a crucial role, including adapting to new technologies to provide an active learning student experience.

Girls and science education: Exploring female interests towards learning with Serious Games a study of KS3 girls in the North East of England

Following the rapid advancement in technologies and computer capabilities, researchers have predicted that more technology-enhanced learning will occur and educational computer games (Serious Games) in particular could play an important part in education. Research has shown that for serious games to be effective as learning tools, they have to provide an engaging experience to the player. However, very few evidence-based research have been carried out to discover the features of games that make it suitable for learning, how the actual learning can be evaluated and more especially, the gender divide that still exists between females and males in the world of gaming. This research aims to determine the educational features necessary to make a serious game effective as a learning tool, develop an assessment framework for the evaluation of serious games as well as determine if some of the educational features and engagement factors are peculiar to females.