Search results

Interaction content is created during online learning interaction for the exchanged information to convey experience and share knowledge. Prior studies have mainly focused on the quantity of online learning interaction content (OLIC) from the perspective of types or frequency, resulting in a limited analysis of the quality of OLIC. Domain concepts as the highest form of interaction are shown as entities or things that are particularly relevant to the educational domain of an online course. The purpose of this paper is to explore a new method to evaluate the quality of OLIC using domain concepts.

This paper proposes a novel approach to automatically evaluate the quality of OLIC regarding relevance, completeness and usefulness. A sample of OLIC corpus is classified and evaluated based on domain concepts and textual features.

Experimental results show that random forest classifiers not only outperform logistic regression and support vector machines but also their performance is improved by considering the quality dimensions of relevance and completeness. In addition, domain concepts contribute to improving the performance of evaluating OLIC.

This paper adopts a limited sample to train the classification models. It has great benefits in monitoring students’ knowledge performance, supporting teachers’ decision-making and even enhancing the efficiency of school management.

This study extends the research of domain concepts in quality evaluation, especially in the online learning domain. It also has great potential for other domains.

The purpose of this paper is to study heterogeneous transfer learning for activity recognition using heuristic search techniques. Many pervasive computing applications require information about the activities currently being performed, but activity recognition algorithms typically require substantial amounts of labeled training data for each setting. One solution to this problem is to leverage transfer learning techniques to reuse available labeled data in new situations.

This paper introduces three novel heterogeneous transfer learning techniques that reverse the typical transfer model and map the target feature space to the source feature space and apply them to activity recognition in a smart apartment. This paper evaluates the techniques on data from 18 different smart apartments located in an assisted-care facility and compares the results against several baselines.

The three transfer learning techniques are all able to outperform the baseline comparisons in several situations. Furthermore, the techniques are successfully used in an ensemble approach to achieve even higher levels of accuracy.

The techniques in this paper represent a considerable step forward in heterogeneous transfer learning by removing the need to rely on instance – instance or feature – feature co-occurrence data.

Connectivism has been proposed to explain the impact of new technologies on learning. According to this approach, learning may occur even outside the individual within an organization or a system. Learning objectives are not defined in advance and learning requires the ability to form connections and use networks to find the required knowledge. The connections by which individuals can learn are more important than what they currently know. The purpose of this paper is to investigate if a measure, rating the importance of concepts, can be derived from a network representation of the learning domain and if highly connected concepts – with high importance value – can describe whether information is explored in such ways as assumed by connectivism.

The authors empirically examined if the proposed measure can provide insight on the role of connections in learning and explain the reasons behind passing certain parts of a test using a linear regression model.

The results are twofold. First, an implementation of the information exploration principle of connectivism has been introduced, applying semantic technologies and the importance measure. Second, although no significant effects could be isolated, trends in performance improvement concerning highly important concepts were identified.

However, connectivism has been known since 2005, it is still lacking for successful implementations. The presented approach of a concept importance measure is a promising starting point by providing means of connected learning, enabling individuals to effectively improve their personal abilities to better fit job demand.

The purpose of this paper is to shed light on the problematics of learning across knowledge boundaries in organizational settings. The paper specifically explores learning processes that emerge, when a new knowledge domain is introduced into an existing organizational practice with the aim of creating a new combined practice.

A case study was carried out as a “natural experiment” in an engineering consultancy, where emerging initiatives to integrate the newly acquired competencies into the existing practice were explored. A theoretical framework informed by selected perspectives on learning processes and boundary processes was applied on three illustrative vignettes to illuminate learning potentials and shortcomings in boundary processes.

In the engineering consultancy, it was found that while learning did occur in the consultancy organization, it remained discrete in ‘pockets’ of learning; mainly at an individual level, at project level or as domain-specific learning. Learning processes were intertwined with elements of domain-specific interests, power, managerial support, structural conditions, material and epistemic differences between knowledge domains.

The finding in this paper is based on a single case study: hence, the findings' generalizability may be limited.

The paper argues that learning across knowledge domains needs various forms of supporting initiatives and constant readiness to alter or counteract when an initiative's shortcomings appear or undesired learning loops arise.

The paper contributes to understanding the complexity of learning across knowledge boundaries in organizational settings.

Approaches to learning have the ability to influence knowledge acquisition, comprehension, retention and even motivation to learn. Previous work indicates that despite age, experience, or prior knowledge, students have a tendency to approach learning differently as a function of the presented content. The purpose of this study is to explore how context influences student approaches to learning science.

The authors adopt a question-asking methodology to evaluate if approaches to learning the same science content vary when presented within the context of Pure Science or the History of Science.

Results indicate that contextualizing the presentation of science content, shifts the approaches students take in attempting to learn science content as evidenced by the questions they ask to deepen their understanding. Additional variables of prior experience with each scientific concept, task persistence at a distractor task and later recall of the presented concepts were related to different inquiry strategies.

Implications for instructional design and pedagogy are discussed.

The framework in which scientific information is presented may impact how students modify existing and create a new schema, impacting their beliefs about scientific knowledge and the way in which students question, hypothesize and engage within the domain of science.

By studying the role of inquiry while students engage in science learning, the authors explore the role of context, content and knowledge retention.

The current study probes at the nature of student questioning and its reliance on the content, context and its relationship to outcome variables such as learning and, perhaps, even persistence as it relates to students’ prior knowledge within content areas which may, in turn, lead to varying levels of student self-efficacy.

Our university demonstrates a strong investment in online education and as part of continuing development delivers some existing online programs in a 3D virtual world. Faculty members need a plan to engage, so they were guided in the adoption of our cybergogy of learning archetypes and learning domains to draw together various aspects of learning. Together we weave threads from orthodox theories with a doctrine of educational technologies that encompasses social-centric 3D interactive virtual environments. This chapter documents the growth of the model from theory into practice to provide a framework for instructors to plan their virtual courses. Five Second Life®-enhanced courses were developed, scheduled and marketed to enrolled students to test the framework. The teaching and learning strategies adopted are reported and outcomes are presented.

The scientific community shares a heritage of knowledge generated by several different fields of research. Identifying how scientific interest evolves is relevant for recording and understanding research trends and society’s demands.

This article presents SciBR-M, a novel method to identify scientific interest evolution from bibliographic material based on Formal Concept Analysis. The SciBR-M aims to describe the thematic evolution surrounding a field of research. The method begins by hierarchically organising sub-domains within the field of study to identify the themes that are more relevant. After this organisation, we apply a temporal analysis that extracts implication rules with minimal premises and a single conclusion, which are helpful to observe the evolution of scientific interest in a specific field of study. To analyse the results, we consider support, confidence, and lift metrics to evaluate the extracted implications.

The authors applied the SciBR-M method for the Educational Data Mining (EDM) field considering 23 years since the first publications. In the digital libraries context, SciBR-M allows the integration of the academy, education, and cultural memory, in relation to a study domain.

Cultural changes lead to the production of new knowledge and to the evolution of scientific interest. This knowledge is part of the scientific heritage of society and should be transmitted in a structured and organised form to future generations of scientists and the general public.

The method, based on Formal Concept Analysis, identifies the evolution of scientific interest to a field of study. SciBR-M hierarchically organises bibliographic material to different time periods and explores this hierarchy from proper implication rules. These rules permit identifying recurring themes, i.e. themes subset that received more attention from the scientific community during a specific period. Analysing these rules, it is possible to identify the temporal evolution of scientific interest in the field of study. This evolution is observed by the emergence, increase or decrease of interest in topics in the domain. The SciBR-M method can be used to register and analyse the scientific, cultural heritage of a field of study. In addition, the authors can use the method to stimulate the process of creating knowledge and innovation and encouraging the emergence of new research.

The study aims to cope with the problems confronted in the skin lesion datasets with less training data toward the classification of melanoma. The vital, challenging issue is the insufficiency of training data that occurred while classifying the lesions as melanoma and non-melanoma.

In this work, a transfer learning (TL) framework Transfer Constituent Support Vector Machine (TrCSVM) is designed for melanoma classification based on feature-based domain adaptation (FBDA) leveraging the support vector machine (SVM) and Transfer AdaBoost (TrAdaBoost). The working of the framework is twofold: at first, SVM is utilized for domain adaptation for learning much transferrable representation between source and target domain. In the first phase, for homogeneous domain adaptation, it augments features by transforming the data from source and target (different but related) domains in a shared-subspace. In the second phase, for heterogeneous domain adaptation, it leverages knowledge by augmenting features from source to target (different and not related) domains to a shared-subspace. Second, TrAdaBoost is utilized to adjust the weights of wrongly classified data in the newly generated source and target datasets.

The experimental results empirically prove the superiority of TrCSVM than the state-of-the-art TL methods on less-sized datasets with an accuracy of 98.82%.

Experiments are conducted on six skin lesion datasets and performance is compared based on accuracy, precision, sensitivity, and specificity. The effectiveness of TrCSVM is evaluated on ten other datasets towards testing its generalizing behavior. Its performance is also compared with two existing TL frameworks (TrResampling, TrAdaBoost) for the classification of melanoma.

Computer-supported collaborative learning (CSCL) is widely used in different levels of education across disciplines and domains. Researchers in the field have proposed various conceptual frameworks toward a comprehensive understanding of CSCL. However, as the definition of CSCL is varied and contextualized, it is critical to develop a shared understanding of collaboration and common definitions for the metrics that are used. The purpose of this research is to present a synthesis that focuses explicitly on the types and features of coding schemes that are used as analytic tools for CSCL.

This research collected coding schemes from researchers with diverse backgrounds who participated in a series of workshops on collaborative learning and adaptive support in CSCL, as well as coding schemes from recent volumes of the International Journal of Computer-Supported Collaborative learning (ijCSCL). Each original coding scheme was reviewed to generate an empirically grounded framework that reflects collaborative learning models.

The analysis generated 13 categories, which were further classified into three domains: cognitive, social and integrated. Most coding schemes contained categories in the cognitive and integrated domains.

This synthesized coding scheme could be used as a toolkit for researchers to pay attention to the multiple and complex dimensions of collaborative learning and for developing a shared language of collaborative learning.

By analyzing a set of coding schemes, the authors highlight what CSCL researchers find important by making these implicit understandings of collaborative learning visible and by proposing a common language for researchers across disciplines to communicate by referencing a synthesized framework.

The present paper introduces concept mapping as a structured participative conceptualization approach to identify clusters of ideas and opinions generated by experts within the domain of mobile learning. Utilizing this approach, the paper aims to contribute to a definition of key domain characteristics by identifying the main educational concepts related to mobile learning.

A short literature review points out the attempts to find a clear definition for mobile learning as well as the different perspectives taken. Based on this an explorative case study was conducted, focusing on the educational problems that underpin the expectations on mobile learning. Using the concept mapping approach the study identified these educational problems and the related domain concepts. The respective results were then analyzed and discussed.

The chosen approach produced several means to interpret the experts' ideas and opinions, such as a cluster map illustrating and structuring substantial accordances. These means help to gain new insights on the emphasis and relation of the core educational concepts of mobile learning. The core educational concepts of mobile learning identified are: “access to learning”, “contextual learning”, “orchestrating learning across contexts”, “personalization”, and “collaboration”.

The paper is original as it uses a unique conceptualization approach to work out the educational problems that can be addressed by mobile learning and thus contributes to a domain definition based on identified issues, featured concepts, and derived challenges. In contrast to existing approaches for defining mobile learning, the present approach relies completely on the expertise of domain experts.