The use of social media in higher education has been demonstrated in a number of studies to be an attractive and contemporary method of teaching and learning. However, further research and investigation are required in order to align social media's pedagogical benefits with the theoretical perspectives that inform educational practices. It is the objective of this study to provide a systematic literature review using bibliometric analysis techniques and content analysis to provide a map of research produced between 2009 and 2021. This study aims to identify theoretical frameworks, current research trends, and patterns in this field. A total of 772 publications were analysed using bibliometric methodology, while a subset of 55 publications were analysed using content analysis. As indicated by the results, there is still a growing interest in this area of research, with recent studies still focusing on attitudes towards the use of social media in teaching and learning. According to the content analysis, technology acceptance theories and learning theories are the most commonly used reference theories. This field has yet to elaborate on pedagogical theory, and there is a tendency to rely primarily on technology acceptance models rather than pedagogical models. A discussion of future practice and research implications is also provided.

The pandemic, that started in the year 2019 due to the SARS-CoV-2 virus, did not interrupt the 3D Modeling & BIM workshop, which has been held telematically ever since. This year, in the edition dedicated to "Information and 3D Modeling for the Built Heritage", I had the pleasure of participating as keynote speaker in the role of researcher and professional expert in visual programming processes for architecture and design. My speech aimed to emphasise how Building Information Modeling (BIM) today is a great container of knowledge, capable of combining the various disciplines that gravitate around the vast theme of modelling, particularly for historical buildings.

We present a novel solution to the attitude control problem of ground vehicles by means of the Active Front Steering (AFS) system. The classical feedback linearization method is often used to track a reference yaw dynamics while guaranteeing vehicle stability and handling performance, but it is difficult to apply because it relies on the exact knowledge of the nonlinearities of the vehicle, in particular the tire model. In this work, the unknown nonlinearities are real-time learnt on the basis of the universal approximation property, widely used in the area of neural networks. With this approximation method, the Uniform Ultimate Boundedness (UUB) property with respect to tracking and estimation errors can be formally proven. Preliminary simulation results show good tracking capabilities when model and parameters are affected by uncertainties, also in presence of actuator saturation.

GOBLET, the Global Organisation for Bioinformatics Learning, Education and Training, is a legally registered foundation whose mission is to cultivate the global bioinformatics trainer community, set standards and provide high-quality resources to support learning, education and training. GOBLET main objectives are: for the training portal to become both a pull mechanism and repository; to offer training for trainers and end-users; to develop training material and course standards; to provide training resources (surveys, best-practice guidelines, etc.); to raise funds to be able to meet our objectives; to offer a network/community forum; to give further consideration to mechanisms for trainer recognition.

In the European Union, the recent update to data protection laws by virtue of the General Data Protection Regulation (GDPR) significantly changed the landscape of the processing of personal data. Consequently, adequate solutions to ensure that the controller and processor properly understand and meet the data protection requirements are needed. In enterprise reality it is quite common to use Business Process (BP) models to manage the different business activities. Hence the idea of integrating privacy concepts into BP models so as to leverage them to the role of GDPR recommenders. To this end, suggestions and recommendations about data management pursuant to GDPR provisions have been added to specific tasks of the BP, to improve both the process management and personnel learning and training. Feasibility of the proposed idea, implemented into an Eclipse plugin, has been provided through a realistic example.

When someone says a statement about a particular subject, we memorize the assertion and, implicitly, we can construct all the possible questions that have as a right answer to the assertion just heard. This means that, in this specific case, our learning process based on assertions subsists. When we read a book, we do nothing but learn through a succession of assertions. In this article, we present a system for automatically constructing a conversational agent, which uses only assertions to build the dialog engine. The whole architecture is based on the "Robot Operating System" (ROS), and the experiments were conducted using a humanoid robot.

There is a burgeoning body of evidence suggesting that technology can enhance learning, and multiple studies have shown that videos represent a very effective tool in science engagement and education. In fact, video media can transform the complexities of science and nature into something more tangible and tractable (Dabylchuk et al., 2018). Documentaries can hence be powerful tools for learning, to raise awareness for important topics, and adding a scientific approach makes it possible to deepen knowledge about the world. According to Dale's cone of experience (Fig. 1), people memorize 10% of what they read, 20% of what they hear, 30% of what they see and 50% of what they hear and see (Wiman and Mierhenry, 1969). These statistics seems to convey a very clear message: blind or visually impaired people are penalised in the learning compared to sighted ones. However, our experience in the area of inclusive science communication has led us to wonder: are we sure that sight is a fundamental means for learning? Or is it possible that sight is a predominant sense over others, and therefore it can even mislead or limit learning? To answer this question, we held an ad-hoc educational laboratory with the users of the "IstitutodeiCiechiOpereRiunite I. Florio - F. ed A. Salamone" of Palermo, with which we have established a convention on 29/01/2019. Specifically, we involved 20 users with different degrees of visual impairment, in the projection of three science videos, selected from those realised by the EDUlab divulgation group over the years. At the end of the video projection, users were asked to express their opinions and their personal interpretations of scientific issues, allowing us to understand what e how many information have been received and, above all, if sighted usersare really advantaged compared to the visually impaired ones.

The detection and representation of events is a critical element in automated surveillance systems. We present here an ontology for representing complex semantic events to assist video surveillance-based vandalism detection. The ontology contains the definition of a rich and articulated event vocabulary that is aimed at aiding forensic analysis to objectively identify and represent complex events. Our ontology has then been applied in the context of London Riots, which took place in 2011. We report also on the experiments conducted to support the classification of complex criminal events from video data.

This Manifesto is one of the final outputs of the Gaming Horizons project (https://www.gaminghorizons.eu/) and summarizes a number of recommendations emerged from previous phases of work by expressing them in terms of 4 Foundational Statements and 9 Actions.

In a joint effort between the University of Amsterdam and ISTI-CNR, researchers have studied the negative impact that low-quality training data (i.e., training data annotated by non-authoritative assessors) has on information extraction (IE) accuracy.

Decision Support Systems (DSSs) based on Fuzzy Logic are gaining increasing research interest in order to solve classification problems in a wide range of application fields, especially in medicine, where the chance of presenting classification results together with a clear explanation and with a measure of the associated uncertainty is highly appealing. However, designing a fuzzy system is a thorny process, requiring many steps to be accomplished, from the knowledge extraction and representation, to the inference process, until the presentation of results. Therefore, this paper proposes a general procedure for constructing rule-based fuzzy classifiers, according to the system characteristics of performance and interpretability required by the specific application, which can be used with any type of data, and is particularly useful for the medical field requirements. The proposed procedure is based on the naïve Bayes approximation, therefore, the optimization of necessary parameters is performed only once and separately for each variable, thus resulting computationally fast, while later steps of the procedure enable to calculate more complicate models and choose the best one, without any further optimization. Moreover, the choices of all degrees of freedom of the design, associated with the variables constituting the model, their fuzzy partitions, the rule base construction, and the inference process, are suggested in this paper. Some of them are motivated by general considerations regarding systems applied in the medical ambit. Some other design choices depend on the dataset and on the application. In order to provide an objective way for choosing these degrees of freedom, some parameters for defining the required trade-off between performance and interpretability are proposed here. The application of the proposed procedure is guided by showing a running example, using data of the Wisconsin Breast Cancer Dataset. For different values of the trade-off parameters, optimal interpretability, or first-rate performance, or acceptable interpretability and performance are obtained, with respect to the best fuzzy systems applied on the same dataset. Finally, the procedure is applied on a number of benchmark datasets, and outstanding results are achieved in terms of performance, with respect to the best classification methods of the state-of-the-art.

In this paper we show how a multilayer neural network trained to master a context-dependent task in which the action co-varies with a certain stimulus in a first context and with a second stimulus in an alternative context exhibits selective attention, i.e. filtering out of irrelevant information. This effect is rather robust and it is observed in several variations of the experiment in which the characteristics of the network as well as of the training procedure have been varied. Our result demonstrates how the filtering out of irrelevant information can originate spontaneously as a consequence of the regularities present in context-dependent training set and therefore does not necessarily depend on specific architectural constraints. The post-evaluation of the network in an instructed-delay experimental scenario shows how the behaviour of the network is consistent with the data collected in neuropsychological studies. The analysis of the network at the end of the training process indicates how selective attention originates as a result of the effects caused by relevant and irrelevant stimuli mediated by context-dependent and context-independent bidirectional associations between stimuli and actions that are extracted by the network during the learning.

This chapter describes a modeling method that has been conceived to support learning in public administrations. The modeling method foresees the description of both procedures in the public administrations, and the working context of the civil servants. The approach relies on several model types that are used to organize and to relate the knowledge needed by civil servants in order to perform their daily activities. Each model instance describes a view on the concerns expressed by the model type it conforms to. These descriptions intend to provide an easy way for civil servants to retrieve knowledge when they need to learn specific aspects of a procedure, and to make collaboration easier in order to enable the emergence of knowledge related to the procedures themselves. Indeed, the method comes with an infrastructure that allows to automatically set up a wiki-based collaborative platform enabling collaboration and knowledge sharing among the stakeholders involved in the activities of a Public Administration. This chapter mainly reports on the modeling method that was conceived and developed within the FP7 EU research project Learn PAd. Learning aspects, while clearly relevant for the project, will not be directly discussed here.

How do humans and other animals face novel problems for which predefined solutions are not available? Human problem solving links to flexible reasoning and inference rather than to slow trial-and-error learning. It has received considerable attention since the early days of cognitive science, giving rise to well known cognitive architectures such as SOAR and ACT-R, but its computational and brain mechanisms remain incompletely known. Furthermore, it is still unclear whether problem solving is a "specialized" domain or module of cognition, in the sense that it requires computations that are fundamentally different from those supporting perception and action systems. Here we advance a novel view of human problem solving as probabilistic inference with subgoaling. In this perspective, key insights from cognitive architectures are retained such as the importance of using subgoals to split problems into subproblems. However, here the underlying computations use probabilistic inference methods analogous to those that are increasingly popular in the study of perception and action systems. To test our model we focus on the widely used Tower of Hanoi (ToH) task, and show that our proposed method can reproduce characteristic idiosyncrasies of human problem solvers: their sensitivity to the "community structure" of the ToH and their difficulties in executing so-called "counterintuitive" movements. Our analysis reveals that subgoals have two key roles in probabilistic inference and problem solving. First, prior beliefs on (likely) useful subgoals carve the problem space and define an implicit metric for the problem at hand-a metric to which humans are sensitive. Second, subgoals are used as waypoints in the probabilistic problem solving inference and permit to find effective solutions that, when unavailable, lead to problem solving deficits. Our study thus suggests that a probabilistic inference scheme enhanced with subgoals provides a comprehensive framework to study problem solving and its deficits.

JaTeCS is an open source Java library focused on automatic text categorization. It covers all the steps of an experimental activity, from reading the corpus to the evaluation of the results. JaTeCS focuses on text as the central input, and its code is optimized for this type of data. As with many other machine learning (ML) frameworks, JaTeCS provides data readers for many formats and well-known corpora, NLP tools, feature selection and weighting methods, the implementation of many ML algorithms as well as wrappers for well-known external software (e.g., libSVM, SVMlight). JaTeCS also provides the implementation of methods related to text classification that are rarely, if never, provided by other ML framework (e.g., active learning, quantification, transfer learning).

JaTeCS is an open source Java library focused on Automatic Text Categorization (ATC). It covers all the steps of an experimental activity, from reading the corpus to the evaluation of the experimental results. JaTeCS focuses on text as the central input, and its code is optimized for this type of data. As with many other machine learning (ML) frameworks, it provides data readers for many formats and well-known corpora, NLP tools, feature selection and weighting methods, the implementation of many ML algorithms as well as wrappers for well-known external software (e.g., libSVM, SVM_light). JaTeCS also provides the implementation of methods related to ATC that are rarely, if never, provided by other ML framework (e.g., active learning, quantification, transfer learning).

Sentiment classification has become a ubiquitous enabling technology in the Twittersphere, since classifying tweets according to the sentiment they convey towards a given entity (be it a product, a person, a political party, or a policy) has many applications in political science, social science, market research, and many others. In this paper, we contend that most previous studies dealing with tweet sentiment classification (TSC) use a suboptimal approach. The reason is that the final goal of most such studies is not estimating the class label (e.g., Positive, Negative, or Neutral) of individual tweets, but estimating the relative frequency (a.k.a. "prevalence") of the different classes in the dataset. The latter task is called quantification, and recent research has convincingly shown that it should be tackled as a task of its own, using learning algorithms and evaluation measures different from those used for classification. In this paper, we show (by carrying out experiments using two learners, seven quantification-specific algorithms, and 11 TSC datasets) that using quantification-specific algorithms produces substantially better class frequency estimates than a state-of-the-art classification-oriented algorithm routinely used in TSC. We thus argue that researchers interested in tweet sentiment prevalence should switch to quantification-specific (instead of classification-specific) learning algorithms and evaluation measures.

Referaggio di un articolo da pubblicare su Cognitive Neurodynamics

While the use of social networks (SNs) and social media is increasingly permeating all sectors of the global society, in Italy there is an ongoing debate about its advantages and drawbacks for learning, especially within formal educational contexts. In order to contribute to such a debate, a study has been conducted, aimed to investigate the Italian university students' beliefs about the positive and negative effects of social networking on their learning and to identify any correlation between such beliefs and the students' characteristics. This chapter reports and discusses the results of the study, which was based on the data collected through a survey to 336 Italian university students (F = 63.6, 83.8 % aged below 32). Results revealed that Italian university students perceive social networks as useful tools for both improving their learning and connecting with their peers, but also that they are aware of their undesirable consequences, such as experiencing negative emotions, losing concentration and being prevented from engaging in extra-academic activities.

Sentiment classification has become a ubiquitous enabling technology in the Twittersphere, since classifying tweets according to the sentiment they convey towards a given entity (be it a product, a person, a political party, or a policy) has many applications in political science, social science, market research, and many others. In this paper we contend that most previous studies dealing with tweet sentiment classification (TSC) use a suboptimal approach. The reason is that the final goal of most such studies is not estimating the class label (e.g., Positive, Negative, or Neutral) of individual tweets, but estimating the relative frequency (a.k.a. "prevalence") of the different classes in the dataset. The latter task is called quantification, and recent research has convincingly shown that it should be tackled as a task of its own, using learning algorithms and evaluation measures different from those used for classification. In this paper we show, on a multiplicity of TSC datasets, that using a quantification-specific algorithm produces substantially better class frequency estimates than a state-of-the-art classification-oriented algorithm routinely used in TSC. We thus argue that researchers interested in tweet sentiment prevalence should switch to quantification-specific (instead of classification-specific) learning algorithms and evaluation measures