Search

In order to adapt to the needs of the times, many enterprises have combined AI (Artificial Intelligence) technology with enterprise management to enhance their competitiveness, which is also essential in the field of education. This article attempted to integrate AI technology into multimedia teaching to improve the teaching quality of teachers and students' understanding of knowledge. The experiment found that AI technology could effectively improve students' understanding of knowledge points and also improve the clarity of knowledge point explanations during the teaching process. It could be seen that using AI intelligent learning systems for multimedia demonstrations could greatly improve the teaching quality of teachers, as well as the learning interest and effectiveness of students. [ABSTRACT FROM AUTHOR]

The methods to control the user's identity during learning in a distance educational environment are explored in the article. In this regard, there is a problem of replacing a registered user with an unscrupulous one. To prevent unauthorized replacement of a registered user, it is proposed to use his keyboard handwriting, which can be formed during user registration in the system. Cluster keyboard handwriting containing information on the data input characteristics of all keyboard symbols is formed to control the user during operation. It is proposed, depending on the length of time the user presses the keys, to calculate the corresponding mean values and standard deviations. Further, the letters are combined into clusters according to the proximity of the obtained statistical characteristics. For clustering, the Kohonen network is used. Preliminary, the mean values and standard deviations are normalized to the interval [0;1]. The formed clusters store the sets of symbols included in them and their statistical characteristics. This method of forming keyboard handwriting makes it possible to verify the user during the entire time he is in the learning environment. The system stores information about the duration of pressing all the keys on the keyboard. Therefore, the decision on the compliance of a user who is in the system with an authorized user is made depending on the correspondence of the current characteristics of the handwriting with the available characteristics. The experimental data that were used to form the cluster keyboard handwriting are presented in the article. [ABSTRACT FROM AUTHOR]

Incident Reporting Systems (IRS) continue to be an important influence on improving patient safety. IRS can provide valuable insights into how to prevent patients from being harmed at the organizational level. But inadequate expectations and misuse, for performance assessment, patient safety measurement or research, have hindered the full IRS potential. Health care organizations need to develop effective strategies built on trust and truth telling to improve the impact of IRS. This requires strategies to address the limited resources to analyse the near-misses or adverse events; avoid the punitive drift through maintaining the anonymity and protective legislation; integrating IRS and avoiding its confusion with mandatory adverse event response systems; training data analysts to focus on the system instead of the individual through a balanced simple taxonomy; combine the analyses at the local level, to reinforce effective and personalized feedback, with the potential of a national or supranational learning platform. [ABSTRACT FROM AUTHOR]

• A large structured and well-annotated corpus of Arabic audio clips collected from a Quran portal. • Systematic empirical comparison of shallow and DL classifiers on Arabic audio clips. • Ability to discriminate authentic audio clips from closely imitated ones. • Performance evaluation of the well-trained classifiers against human experts. • Superior performance in discriminating authentic Cantillations from imitations. This paper introduces a thorough study of classical and deep learning algorithms implemented for multi-class speaker identification and verification of Qur'anic audio clips. Thirty different reciters and twelve imitators, of top reciters, were evaluated in the study. In addition to identifying the reciter, our objective is to first evaluate different classifiers performing the stated recognition, compare classical vs. deep-based classifiers, and finally benchmark automatic against human expert listeners' accuracy in identifying authentic reciters from imitators. Using different multimedia outlets over the internet, several reciters became more popular than others for their distinct cantillation style. Towards the development of a practical classifying system, a significant dataset of 15810 audio clips is constructed for thirty reciters in addition to 397 clips for top imitators. A combination of perceptual and acoustic features was extracted in order to achieve better classification. The classifying system is implemented using six top-achieving classical and two deep learning classifiers. Finally, we perform a survey engaging human expert listeners' performance in detecting imitators from authentic reciters, against the fine-tuned classifiers to be able to discuss cross-comparative results. The concluded results demonstrated a high accuracy performance for the selected classifiers averaging 98.6 % on the testing dataset, 93 % on a separate testing dataset, 97 % on the problem of discriminating authentic Cantillations from imitations, and outstanding performance on the survey reaching an accuracy rate of 98 % compared to 61 % of human experts. [ABSTRACT FROM AUTHOR]

Due to traffic conditions that are very context dependent, the computation of optimized or shortest paths is a very complex problem for both drivers and autonomous vehicles. In this paper, we introduce a learning mechanism that is able to efficiently evaluate path durations based on an abstraction of the available traffic information. We demonstrate that a cache data structure allows a permanent access to the results whereas a lazy politics taking new data into account is used to increase the viability of those results. Our measures highlight the performance of each mechanism, according to different learning strategies. [ABSTRACT FROM AUTHOR]

Abstract Mobile technologies and their applications have the potential to benefit various learning contexts. Users' perceptions of mobile learning (m-learning) technologies are of great importance and precede the successful integration of these technologies in education. M-learning adoption has been investigated in the literature with reference to various factors and learning analytics, but largely without considering the role of different configurations (i.e., specific combinations of variables), and how these configurations might affect the adoption of various user groups. For instance, users with different backgrounds, experiences, learning styles, and so on might not be represented by the one-model-fits-all produced from the common regression approaches. In this study, we briefly review factors that have been proven important in the context of mobile learning adoption, and build on complexity theory and configuration theory in order to explore the causal patterns of factors that stimulate the use of mobile learning. To test its propositions, the study employs fuzzy-set qualitative comparative analysis (fsQCA) on a data sample from 180 experienced m-learning users. Findings indicate eight configurations of cognitive and affective characteristics, and social and individual factors, that explain m-learning adoption. This research study contributes to the literature by (1) offering new insights on how predictors of m-learning adoption interrelate; (2) extending existing knowledge on how cognitive and affective characteristics, and social and individual factors, combine to lead to high m-learning adoption; and (3) presenting a step-by-step methodological approach for how to apply fsQCA in the area of learning systems and learning analytics. Highlights • How technology acceptance research can capture complex multidimensional phenomena. • How predictors of m-learning adoption interrelate to form complex configurations. • Eight different solutions can explain high m-learning adoption. • A methodological approach on how to apply fsQCA in learning systems and analytics. [ABSTRACT FROM AUTHOR]

Abstract In this paper a novel training technique is proposed to offer an efficient solution for neural network training in non-trivial and critical applications such as the diagnosis of health threatening illness. The presented technique aims to enhance the generalization capability of a neural network while preserving its sensitivity and precision. The implemented method has been devised in order to slowly increase, during training, the generalization capabilities of a Radial Basis Probabilistic Neural Network classifier, as well as preventing it from over-generalization and the consequent lack of resulting classification performances. The developed method was tested on Electrocardiograms. These latter are generally considered non-trivial both due to the difficulty to recognize some anomalous heart activities, and due to the intermittent nature of abnormal beat occurrences. The implemented training method obtained satisfactory performances, sensitivity and precision while showing high generalization capabilities. [ABSTRACT FROM AUTHOR]

Abstract This work provides a rigorous framework for studying continuous-time control problems in uncertain environments. The framework models uncertainty in state dynamics as a probability measure on the space of functions. Such a probability measure is permitted to change over time as agents learn about their environment. This model can be seen as a variant of either Bayesian reinforcement learning (RL) or adaptive optimal control. We study conditions for locally optimal trajectories within this model, in particular deriving an appropriate dynamic programming principle and Hamilton–Jacobi equations. Some discussion of variants of the model are also provided, including one potential framework for studying the tradeoff between exploration and exploitation in RL. [ABSTRACT FROM AUTHOR]

Quality measurement in pediatrics is challenged by a system that lacks a fundamental data infrastructure for children's healthcare in general and in particular in cardiomyopathy. We suggest that the thoughtful application of mixed health services research methods can serve as powerful tools for applied research that supports a priori thinking, which in turn can drive both prospective studies and the analyses of retrospectively collected data within the schemas of strong quasi-experimental designs. This can provide the means for transforming practice into evidence, and practice within uncertainty into deep knowledge. The process of learning is iterative and typically incremental, constantly being infused by every day work experience and hard-earned lessons by clinicians providing clinical care. Developing sustainable learning towards improved outcomes in pediatric cardiomyopathy can be done using the applied science of health services research to translate clinical practice into research. [ABSTRACT FROM AUTHOR]

Quality improvement (QI) is becoming a central part of the work of clinicians throughout healthcare. Continuous quality improvement (CQI), Lean Management Systems (LMS) and Lean Six Sigma (LSS) are management philosophies as well as management methods. They offer an approach, a set of tools, and a way of thinking about how to more effectively assess and study clinical flow, including addressing variation in clinical process and operations. We define CQI as the daily use of QI methods as a regular part of practice engaging all practice staff, constantly measuring structure, processes, outcomes against effective practices (benchmarking), moving from one QI project to the next, pursuing the goal of “The right care for every child every time”. It is based on clear scientific principles, a valid way of measuring change and has theories of reliability and human factors that underpin the interventions. Significant variations in quality of care provided to pediatric patients leading to substandard care have been well documented. For example, in antibiotic prescribing for community acquired pneumonia; pediatric secondhand smoke reduction; screening for diabetes in cystic fibrosis program; and, in depression screening in Type 1 Diabetes. Despite this, not enough pediatric practices are performing continuous quality improvement (CQI) as part of their daily activities. All health care professionals caring for children should consider incorporating quality measurement into their practice. However, we need to focus on what is the right approach to take and the right questions to ask and address the challenges of aggregating scientifically imperfect tests of change. Increasingly, colleagues, patients, payers and certifying agencies expect such measurement to achieve the Triple Aim of better health, better care and lower cost. In addition, new payment models increasingly pay providers for demonstrated value rather than volume, and they expect participation in continuous improvement. Quality Improvement (QI) is a critical component of the American Board of Pediatrics (ABP) Maintenance of Certification (MOC) Part 4. This requires pediatricians to participate in a meaningful manner in two data-driven QI projects every five years. Pediatricians can select quality measures to evaluate whether patient outcomes and experience improve, and if not, identify and overcome barriers. In this paper we discuss performance improvement using CQI and related methods, suggest approaches to help pediatric cardiologists to ask the right questions when seeking to drive improvement, and consider the implications of measurement theory and complexity science for QI and CQI. [ABSTRACT FROM AUTHOR]

In this paper, we aim to tackle the neural robust tracking control problem for a class of nonlinear systems using the adaptive critic technique. The main contribution is that a neural-network-based robust tracking control scheme is established for nonlinear systems involving matched uncertainties. The augmented system considering the tracking error and the reference trajectory is formulated and then addressed under adaptive critic optimal control formulation, where the initial stabilizing controller is not needed. The approximate control law is derived via solving the Hamilton–Jacobi–Bellman equation related to the nominal augmented system, followed by closed-loop stability analysis. The robust tracking control performance is guaranteed theoretically via Lyapunov approach and also verified through simulation illustration. [ABSTRACT FROM AUTHOR]

The paper studies the choosing mechanism of an energy company which gathers a library of electric load models and at every day chooses the best one for daily prediction. We use a combination of a semi-Markov process and a modified hidden Markov chain to describe the joint curve of loads, the daily best model, and exogenous information, of which temperature is an important factor. By extending the state space of the semi-Markov process, then in a computationally tractable way, the problem is embedded within a hidden Markov chain. Hence we can establish an EM algorithm and an enhancing statistical learning method for estimating parameters and forecasting load. Simulation reveals the range in which the proposed algorithm is applicable. Examples from real world datasets show that the proposed automated system is an alternate method for short term electric load forecasting with loads greater than a few hundreds MW. Supplementary material includes scripts of the proposed system and a guide of the scripts. [ABSTRACT FROM AUTHOR]

Human-in-the-loop (HITL), a key branch of Human–Computer Interaction (HCI), is increasingly proposed in the research literature as a key assurance method for automated analyses and predictive application designs. As the need increases to improve methods in Artificial Intelligence (AI) model training, optimize systems performance, provide AI explainability, and monitor AI system operations, the concept of HITL is gaining traction due to its value in solving these challenges. This survey of existing works on HITL from a computer vision system design perspective focuses on the following AI assurance principles: (1) improved data assurance, such as data preparation or automated data labeling; (2) algorithmic assurance, such as managing uncertainty and AI trustworthiness; and (3) critical limitations and capabilities introduced by HITL into a system's operational efficiency. We survey prior work within these foci, including technical strengths and weaknesses of novel approaches and ongoing research. This review of the state of the art in HITL computer vision research supports an informed discussion of considerations and future opportunities in this critical space. • New AI-driven trends are manifesting in the field of HITL. • A convergence of new data and algorithmic assurance methods for HITL is emerging. • Evidence shows that human intervention is necessary for trustworthy computer vision. • State-of-the-art HITL methods are presented and discussed. [ABSTRACT FROM AUTHOR]

Speaking of training, in most cases it comes to the e-learning systems, and how to present the material. However, from the perspective of the authors, not less important is the question: “What to teach?” In order to form an objective idea of the needs of the market the internet service has been developed, development options of which will be considered in this article both from a technical and conceptual perspectives. [ABSTRACT FROM AUTHOR]

The online bin packing problem is a well-known bin packing variant and which requires immediate decisions to be made for the placement of a lengthy sequence of arriving items of various sizes one at a time into fixed capacity bins without any overflow. The overall goal is maximising the average bin fullness. We investigate a ‘policy matrix’ representation, which assigns a score for each decision option independently and the option with the highest value is chosen, for one-dimensional online bin packing. A policy matrix might also be considered as a heuristic with many parameters, where each parameter value is a score. We hence effectively investigate a framework which can be used for creating heuristics via many parameters. The proposed framework combines a Genetic Algorithm optimiser, which searches the space of heuristics in policy matrix form, and an online bin packing simulator, which acts as the evaluation function. The empirical results indicate the success of the proposed approach, providing the best solutions for almost all item sequence generators used during the experiments. We also present a novel fitness landscape analysis on the search space of policies. This study hence gives evidence of the potential for automated discovery by intelligent systems of powerful heuristics for online problems; reducing the need for expensive use of human expertise. [ABSTRACT FROM AUTHOR]

This paper provides an innovative segmentation approach stemming from the combination of cluster analyses and fuzzy learning techniques. Our research provides a real case solution in the Spanish energy market to respond to the increasing number of requests from industry managers to be able to interpret ambiguous market information as realistically as possible. The learning stage is based on the segments created from a non-hierarchical cluster analysis. This results in fuzzy segmentation which permits patterns to be assigned to more than one segment. This in turn reveals that “fuzzifying” an excluding attitudinal segmentation offers more interpretable and acceptable results for managers. Our results demonstrate that 30% of the individuals show plural patterns of behaviour because they have a significant degree of adequacy to more than one segment. In such a rational market, this fact enables sales forces to develop more precise approaches to capture new customers and/or retain existing ones. [ABSTRACT FROM AUTHOR]

Abstract: As a part of smart homes, a subsystem consisting of three components including a neural network is designed to provide personalized services. Unique factor combinations of building specifics, user profiles and external influences lead to the necessity of self-adaptive systems for personal comfort. The system supports room temperature control in order to heat rooms energy- efficiently at a set time. Smart home systems require a software architecture that allows services to be deployed on virtual and hardware devices.The design of automated processes is the first step of later programming and implementation into smart home systems that will automatically supervise and re-train its components and will also allow live feedback. [Copyright &y& Elsevier]

Abstract: Using new game-based learning systems in college education is neither an easy nor a simple task. The aim of such systems is to keep attention, teach students or assess their knowledge through a game. With the aim of keeping students’ attention through a game, in this paper we show the implementation of game-based learning systems with a pedagogical agent. We presents two models for assessing student’s knowledge used by a pedagogical agent which is a part of the new class of Multimedia Interactive Modules for Learning – MIMLE. One of the models is used for activating the agent. It is realized as a window of Help option and built in accordance to Marcov decision process theory (MDP). The basic goal of this mode is to determine the minimal intervention of the agent towards making the right direction concerning the studying process based on simulation learning. With the second, long-term model, we have assessed student’s knowledge in the current game level that is used to decide students should pass on to the next level of learning or if they should stay on the same level. [Copyright &y& Elsevier]

Abstract: Huge amount of information is available in un-structured (text) documents. Knowledge discovery in un-structured document has been recognized as promising task in the recent years. Since un-structured document is typically formatted for human viewing, it varies widely from document to document. Frequent changes made to their formatting further causes difficulty in construction of a global schema. So, Discovery of interesting rules form it is complex and tedious process. Most of the existing system uses hand-coded wrappers to extract information, which is monotonous and time consuming. In this paper we propose a novel and hybrid approach of learning (context-free) grammar rules that are based on alignment between texts. Also it automatically discovers the grammar rules using grammatical inference of repeated pattern present in un-structured (text) document. The generated rules can be used to infer the attribute value pairs from the unstructured text document. [Copyright &y& Elsevier]

Abstract: Nowadays, processing all information of a fusion database is a much more important issue than acquiring more data. Although typically fusion devices produce tens of thousands of discharges, specialized databases for physics studies are normally limited to a few tens of shots. This is due to the fact that these databases are almost always generated manually, which is a very time consuming and unreliable activity. The development of automatic methods to create specialized databases ensures first, the reduction of human efforts to identify and locate physical events, second, the standardization of criteria (reducing the vulnerability to human errors) and, third, the improvement of statistical relevance. Classification and regression techniques have been used for these purposes. The objective has been the automatic recognition of physical events (that can appear in a random and/or infrequent way) in waveforms and video-movies. Results are shown for the JET database. [ABSTRACT FROM AUTHOR]

Abstract: Fisheries are complex human-in-nature systems. The conventional approach to fisheries systems has been to treat them as predictable and controllable. As complex systems they are neither of the two and have to be approached differently. Complex systems often exhibit the capacity to self-organize or adapt, even without outside influence. If this is true of fisheries, it should lead to a radically different approach to management of fisheries systems that places much emphasis on enabling self-organization, learning and adaptation. Conceptual and practical frameworks for enabling activities are needed. [Copyright &y& Elsevier]

Abstract: The conceptual framework for this study was based on the premise that there are lifestyle groups within a culture which commonly share attitudes and perceptions which are reflected in the form and design of their buildings. This study focussed on a particular lifestyle group of people from the same ethnic background, i.e. Caucasian of European ancestry who owned and lived in detached houses. The sample of participants were homogenous in terms of their ethnic background and houses. The setting of this study was a residential estate in Townsville, Queensland, Australia, a hot-humid environment. This study used the ethnographic approach of case study methodology; hence, data were gathered from questionnaires, interviews, observations, building plans, notes and photographs. The attitudes and perceptions of the selected group of homeowners on the quality of their houses in six areas of building performance: spatial, thermal, acoustical, visual, building integrity and indoor air quality, were investigated. The data were qualitatively analysed, assembled and manipulated in a computer-based model. In developing the computer-based model, the capabilities of the programming paradigms—conventional programs, knowledge-based systems and learning systems, were investigated. A neural network learning system was chosen for developing the computer-based model. The neural network model was tested and validated on example cases of houses to verify its prediction capability. [Copyright &y& Elsevier]

Abstract: In this paper, we present a vessel enhancement method, SVM temporal filtering (STF), for X-ray angiographic (XA) images using Support Vector Machine (SVM). We show that the linear SVM applied to vessel enhancement can be regarded as a matched linear filter optimizing the contrast-to-noise ratio in XA images. We propose a non-linear kernel function for the SVM leading to good enhancement with noisy, varying grey-level dynamics at vessel pixels. One key advantage over the matched filters is that an optimal filter is learnt from images, not estimated at design stage. Results on clinical XA images show that learning-based enhancement achieves better results compared to simple subtraction and other image stacking methods. [Copyright &y& Elsevier]

Abstract: Modeling capabilities of two types of learning systems are compared: the naïve Bayesian classifier (NBC) and artificial neural networks (ANNs), based on their prediction errors and relative importance factors of input signals. Simulated and real industrial data were used. It was found that NBC can be an effective and, in some applications, a better tool than ANNs. [Copyright &y& Elsevier]

Globalization and growing new markets, as well as increasing emission and fuel consumption requirements, force the car manufacturers and their suppliers to develop new engine control strategies in shorter time periods. This can mainly be reached by development tools and an integrated hardware and software environment enabling rapid implementation and testing of advanced engine control algorithms.The structure of a rapid control prototyping (RCP) system is explained, which allows fast measurement signal evaluation, and rapid prototyping of advanced engine control algorithms. A hardware-in-the-loop simulator for diesel engine control design is illustrated, simulation results for a 40 tons truck are presented. Providing efficient engine models for the proposed development tools, a dynamic local linear neural network approach is explained and applied for modelling the NOx emission characteristics of a 1.9 l direct injection diesel engine. Furthermore the application of a RCP system is exemplified by the application of combustion pressure based closed-loop ignition timing control for a SI engine. Experimental results are shown for a 1.0 l SI engine on a dynamic engine test stand. [Copyright &y& Elsevier]

Rapid growth of timber, mining and ranching industries in forested areas worldwide often offer small holders opportunities to sell forest resources. Rural communities, however, often have little notion of the market value or economic and ecological consequences of forest transformation. Within such scenarios, the learning process needs to be consciously constructed so as to catalyse new ways of thinking about forest management effectively and quickly. This article describes an ecological research project that integrated data and process-oriented approaches to promote collaborative learning. Results indicate that user-centered approaches are needed to ensure that locally relevant information is generated by scientists, and that learning is catalysed not only among information saturated stakeholders such as policy makers and academics, but also among stakeholders who are directly dependent upon forest resources. [Copyright &y& Elsevier]

Time series classification (TSC) is the problem of categorizing time series data by using machine learning techniques. Its applications vary from cybersecurity and health care to remote sensing and human activity recognition. In this paper, we propose a novel multi-process collaborative architecture for TSC. The propositioned method amalgamates multi-head convolutional neural networks and capsule mechanism. In addition to the discovery of the temporal relationship within time series data, our approach derives better feature extraction with different scaled capsule routings and enhances representation learning. Unlike the original CapsNet, our proposed approach does not need to reconstruct to increase the accuracy of the model. We examine our proposed method through a set of experiments running on the domain-agnostic TSC benchmark datasets from the UCR Time Series Archive. The results show that, compared to a number of recently developed and currently used algorithms, we achieve 36 best accuracies out of 128 datasets. The accuracy analysis of the proposed approach demonstrates its significance in TSC by offering very high classification confidence with the potential of making inroads into plentiful future applications. [ABSTRACT FROM AUTHOR]

• A new deep learning framework is proposed in this paper, which can leverage the object label inter-dependent for weakly supervised learning. • We creatively introduce a novel initialization method of label embeddings for inter-relationships learning. • Ablation studies of our model are provided in detail. • Our models can be applied to object classification and weakly supervised pointwise object localization. In computer vision, the research community has been looking to how to benefit from weakly supervised learning that utilizes easily obtained image-level labels to train neural network models. The existing deep convolutional neural networks for weakly supervised learning, however, generally do not fully exploit the label dependencies in an image. To make full use of this information, in this paper, we propose a new framework for weakly supervised learning of deep convolutional neural networks, introducing graph convolutional networks to capture the semantic label co-occurrence in an image. Moreover, we propose a novel initialization method for label embedding in graph convolutional networks, which enables a smoother optimization for interrelationships learning. Extensive experiments and comparisons on four public benchmark datasets (PASCAL VOC 2007, PASCAL VOC 2012, Microsoft COCO, and NUS-WIDE) show the superior performance of our approach in both image classification and weakly supervised pointwise object localization. These results lead us to conclude that the label dependencies in the input image can provide valuable evidence for learning strongly localized features. [ABSTRACT FROM AUTHOR]

'Safety-II' is a new approach to safety, which emphasizes learning proactively about how safety and efficacy are achieved in everyday frontline work. Previous research developed a new lesson-sharing tool designed based on the Safety-II approach: Resilience Engineering Tool to Improve Patient Safety (RETIPS). The tool comprises questions designed to elicit narratives of adaptations that have contributed to effectiveness in care delivery. The purpose of this study is to revise and validate the design of RETIPS. The tool was revised based on feedback of clinicians at a large multi-specialty hospital, resulting in a version customized for anesthesia residents, RETIPS-AnRes. RETIPS-AnRes was administered on a pilot-basis to anesthesia resident groups for a limited period of time. A review of the reports obtained shows a strong alignment of responses with the conceptual basis of the tool, i.e. learning about how things go well in everyday work. The exemplars include both, specific instances of successful patient care, as well as generic routines that contribute to safe and/or effective care delivery. These findings support RETIPS as a tool to operationalize the Safety-II paradigm in healthcare. Lessons and implications for implementation on a wider scale are discussed. [ABSTRACT FROM AUTHOR]

• Generation of a common knowledge regarding u-learning situation awareness. • Distinction between u-learning situation awareness process and product. • Organisation of the generated knowledge on 4 structured and complementary views. • Studied views: Product view, Process view, Representation view and Enactment view. • Evaluation and discussion of previous research works given the framework's views. This research is devoted to the study of the situation awareness from ubiquitous learning (u-learning) systems' perspective. Related researches identified literature provide different visions and sometimes conflicting ones to the concept of situation awareness. Given this misunderstanding, this paper aims to bring a common and organized characterisation of this concept for the u-learning field. A multidimensional framework dedicated for situation-aware u-learning systems is then defined. This framework is composed of four different and complementary worlds. Each world helps to understand, analyse and describe in-depth a particular aspect of situation awareness in u-learning systems. The paper presents a structured description of the framework's worlds. An evaluation of the research works identified in literature based on the framework's worlds is then presented. The provided assessment helps identifying problems involving situation awareness in u-learning systems which could be a starting point for possible future research opportunities. [ABSTRACT FROM AUTHOR]