Traditionally, e-Learning content is delivered without taking the learner’s traits into account. Content delivered to the learner should be personalized based on the learner profile so that learning can be effective. Also, assessment of a learner’s learning objective is normally done by posing a set of questions without documenting the student’s capabilities. A school of thought envisages asses...

Though deep neural networks have shown great success in the large data domain, they generally perform poorly on few-shot learning tasks, where a classifier has to quickly generalize after seeing very few examples from each class. The general belief is that gradient-based optimization in high capacity classifiers requires many iterative steps over many examples to perform well. Here, we propose ...

Bayesian Belief Networks provide a principled, mathematically sound, and logically rational mechanism to represent student models. The belief net backbone structure proposed by Reye [14,15] offers a practical way to represent and update Bayesian student models describing both cognitive and social aspects of the learner. Considering students as active participants in the modelling process, this ...

This paper presents a new semantics for inductive empirical knowledge. The epistemic agent is represented concretely as a learner who processes new inputs through time and who forms new beliefs from those inputs by means of a concrete, computable learning program. The agent’s belief state is represented hyper-intensionally as a set of time-indexed sentences. Knowledge is interpreted as avoidanc...

We propose a novel approach for credit card fraud detection, which combines evidences from current as well as past behavior. The fraud detection system (FDS) consists of four components, namely, rule-based filter, Dempster–Shafer adder, transaction history database and Bayesian learner. In the rule-based component, we determine the suspicion level of each incoming transaction based on the exten...

The study aims to explore the interrelationships between translation learning belief, strategy use, engagement, anxiety, and their effects on performance at higher education level in China. We administered a questionnaire 339 undergraduates (148 males 191 females). Results showed that: (a) female students had more positive beliefs, as well test scores than male peers; (b) high English possessed...

This paper describes a method for learner modelling for use within simulation-based learning environments. The goal of the learner modelling system is to provide the learner with advice on discovery learning. The system analyzes the evidence that a learner has generated for a specific hypothesis, assesses whether the learner needs support on the discovery process, and the nature of that support...

Learner preferences of students have been analysed within two modules that have been given in the department of media using Röll’s learner preference model. A qualitative analysis of the results of an online learner preference test based on this model has been conducted. This analysis showed that for the majority of the students the revealed results were valid. Starting from current methods use...

Abstract I We propose a new framework for surprise-driven learning that can be used for modeling how humans and animals learn in changing environments. It approximates optimal Bayesian learner, but with significantly reduced computational complexity. I This framework consists of two components: (i) a confidence-adjusted surprise measure to capture environmental statistics as well as subjective ...

An answer of a learner can be interpreted as a learning evidence for demonstrating the understanding of the learner, while a confidence on the answer represents the belief of the learner as the degree of understanding. In this paper, we propose Kit-Build concept map with confidence tagging. KitBuild concept map (KB map in short) is a digital tool for supporting a concept map strategy where lear...