Abstract In recent years, deep learning-based emotion recognition from electroencephalography (EEG) signals has garnered significant attention in brain-computer interfaces. However, effectively capturing local and global dependencies remains a challenge due to the complexities of EEG data. Furthermore, traditional convolutional neural networks and RNNs often struggle to fully explore the spatio-temporal relationships between different features. To address these issues, we propose an end-to-end model with the augmented capsule-gated Transformer to improve the performance of EEG emotion recognition, in which we learn cross-channel spatial features effectively, and the raw EEG signals are automatically weighted to emphasize key attributes. Subsequently, the capsule network extracts low-level and high-level spatial information, fully leveraging the potential insights within the signals. Building on this, an efficient Transformer is employed to model the relationships among different electrodes, allowing for a more in-depth analysis of the temporal dependencies across multiple features. Extensive experiments are conducted on the Dataset for Emotion Analysis using Physiological Signals (DEAP) dataset, and comparison results with existing state-of-the-art methods demonstrate the superior performance of the proposed method. Specifically, for the arousal and valence dimensions, the average recognition accuracies in subject-dependent experiments reach 93.51% and 94.24%, while the subject-independent experiments achieve average accuracies of 86.78% and 87.59%.

The Internet of Things (IoT) paradigm is evolving and the Next-Generation IoT (NG-IoT) ecosystem will incorporate distributed ledger and blockchain technology, AI-adapted components, and intelligent edge solutions that take advantage of edge computing, Artificial Intelligence (AI), networks, and communications. In addition to the low integration of eXplainable Artificial Intelligence (XAI) in the IoT or NG-IoT contexts, the explainability of these systems is rarely evaluated. Due to these limitations, we thoroughly examined the current state of XAI integration with IoT services. We propose a new conceptual framework called eXING-IoT (eXplainability Integrated in the Next Generation IoT) for better NG-IoT systems' explainability integration and evaluation. This includes a list of qualities that future NG-IoT environments should have, thus paving the way for the advancement of NG-IoT beyond the state of the art.

Federated Learning (FL) is a rather new distributed machine learning paradigm based on a collaboratively decentralized privacy-preserving technology. It supports a range of multiple clients (starting from “simple” mobile devices but also including organizations, institutions, etc.) coordinated decentralized machine learning by one or more central servers. As it is a rather new approach various strategies for organization of multiple clients and implementation reliable environments have been developing. The efficiency of selected FL strategy for a particular problem is influenced by involved actors and organizational structure. Particular attention within FL should be paid to adoption of data privacy and security aspects. FL is powerful and widely applicable in areas like banking and finances, automotive industry, IoT and smart environments, health and medicine etc.. In this paper we will pay attention to several crucial aspects of FL, present its essential characteristics, and briefly illustrate several characteristic applications.

In our research, we present a data mining model for the early diagnosis of hyperinsulinemia, potentially reducing the risk of diabetes, heart disease, and other chronic conditions. The dataset, gathered from 2019 to 2022 by Serbia's Healthcare Center through an observational cross-sectional study, includes 1008 adolescents. Medical datasets are often highly imbalanced and may contain irrelevant features that hinder predictive performance. To address these challenges in the medical data analysis, we propose a model employing Functional Principal Component Analysis (FPCA), which also accounts for outliers that could otherwise lead to the inclusion of irrelevant features. Unlike standard Principal Component Analysis (PCA), which is sensitive to the initial positions of cluster centers influencing the final outcome, our model integrates FPCA with K-Means clustering to improve the preprocessing stage. Additionally, we have incorporated the post-hoc explanatory method SHAP (SHapley Additive exPlanations) alongside algorithms such as Random Forest, XGBoost, and LightGBM to provide deeper insights into our model, identifying the most contributory features for the development of hyperinsulinemia. Experimental results showed that combining FPCA with K-Means clustering enhances the accuracy of the XGBoost classifier, with this model achieving an accuracy score of 0.99.

Software estimation rates are still suboptimal regarding efficiency, runtime, and the accuracy of model predictions. Graph Neural Networks (GNNs) are complex, yet their precise forecasting reduces the gap between expected and actual software development efforts, thereby minimizing associated risks. However, defining optimal hyperparameter configurations remains a challenge. This paper compares state-of-the-art models such as Long-Short-Term-Memory (LSTM), Graph Gated Neural Networks (GGNN), and Graph Gated Sequence Neural Networks (GGSNN), and conducts experiments with various hyperparameter settings to optimize performance. We also aim to gain the most informative feedback from our models by exploring insights using a post-hoc agnostic method like Shapley Additive Explanations (SHAP). Our findings indicate that the Taguchi orthogonal array optimization method is the most computationally efficient, yielding notably improved performance metrics. This suggests a compromise between computational efficiency and prediction accuracy while still requiring the lowest number of runnings, with an RMSE of 0.9211 and an MAE of 310.4. For the best-performing model, the GGSNN model, within the Constructive Cost Model (COCOMO), Function Point Analysis (FPA), and Use Case Points (UCP) frameworks, applying the SHAP method leads to a more accurate determination of relevance, as evidenced by the norm reduction in activation vectors. The SHAP method stands out by exhibiting the smallest area under the curve and faster convergence, indicating its efficiency in pinpointing concept relevance.

In contemporary society people constantly are facing situations that influence appearance of serious diseases. For the development of intelligent decision support systems and services in medical and health domains, it is necessary to collect huge amount of patients’ complex data. Patient’s multimodal data must be properly prepared for intelligent processing and obtained results should be presented in a friendly way to the physicians/caregivers to recommend tailored actions that will improve patients’ quality of life. Advanced artificial intelligence approaches like machine/deep learning, federated learning, explainable artificial intelligence open new paths for more quality use of medical and health data in future. In this paper, we will focus on presentation of a part of a novel Open AI Architecture for cancer patients that is devoted to intelligent medical data management. Essential activities are data collection, proper design and preparation of data to be used for training machine learning predictive models. Another key aspect is oriented towards intelligent interpretation and visualisation of results about patient’s quality of life obtained from machine learning models. The Architecture has been developed as a part of complex project in which 15 institutions from 8 European countries have been participated.

Current society is facing remarkable changes in last decade directed by several processes: gains in life expectancy leading to ageing population; essential transformation caused by different epidemic situations including the recent COVID-19 pandemic; on-line work and isolation that leads to different levels of mental disorders; stressful working environments and everyday life. All these circumstances generate significant risk factors that drastically influence peoples' bad health status and conditions.On the other hand, in a lot of, even highly developed countries, health and medical support is getting more and more limited and out of normal working conditions due to different reasons.So, a very important question in this highly technologically developed society is if and how information communication technologies, artificial intelligence, and virtual agent technologies can help in changing and improving the current situation.One important scientific and research direction is oriented towards the development of high quality and reliable personalized medical and health services. The development of sophisticated, intelligent virtual agents i.e. a specific kind of medical e-coaching can help people in getting adequate advices and recommendations in order to improve their health conditions and quality of key life indicators.The use of agents in personalized and social medical and health platforms can open new possibilities for producing tailored recommendations during human-intelligent virtual agents' conversation and support. Some opportunities and challenges of human-agent communication are considered in order to increase human living and health conditions. Several paradigmatic cases of intelligent virtual agents are presented and challenges for future development.

Abstract Rapid and accurate assessment of software project development using artificial intelligence tools can be essential for success in the software industry. This article has two objectives: to reduce the magnitude relative error (MRE) value in estimating the effort and cost of software development using the proposed artificial neural network architecture based on the Taguchi method and examine the influence of input variables on the change in relative error value. Clustering and fuzzification methods further mitigate the heterogeneous structure of the different project values of the datasets used. Taguchi method contributes to the reduction of the number of iterations by 99%, which achieves a significant reduction in estimation and value of MRE. By monitoring additional criteria, such as prediction, correlation, and comparing two activation functions, such as sigmoid and radial basis function, the proposed model's correctness, reliability, and stability are confirmed. Significantly better results are expected using the sigmoid activation function and a decrease in the value of the mean (MRE).

The multidisciplinary nature of intelligent systems creates new challenges every day for researchers but also for companies all over the world. People face such systems in a lot of domains and real-life environments. Not only the development of theoretical aspects, but also practical applications have increasingly spread in the last decade to every field, from health to security or from agriculture to business management. Artificial intelligence, the development of novel, highly innovative theoretical methods and approaches, and also sophisticated applications and services are a focus of both academia and industry. Among numerous applications of artificial intelligence, image processing attracts great attention with the possibility of significant applications in a variety of domains. The combination of innovative algorithms to process massive image data are among the most important focal points of intelligent systems. In addition, innovative approaches in intelligent systems require discovering new problem-solving strategies, models, methodologies, and concurrent algorithms. This special issue of Concurrency and Computation: Practice and Experience includes extended high-quality papers presented at the IEEE International Conference on Innovations in Intelligent Systems and Applications (INISTA), which was held on August 24–26, 2020 in Novi Sad, Serbia, http://inista.org/inista20/index.php. Since 2004, the series of INISTA conferences has focused on intelligent systems, including both software and hardware, and provides a forum for researchers and industry to discuss new ideas and to exchange experiences with particular focus on innovative aspects and applications. INISTA 2020 attracted many international participants with over 80 submissions from all continents. Among them, 51 papers were accepted and published in the IEEE proceedings. Five papers have been selected and invited for this special issue. Authors were asked for extended papers containing at least 40% new material. After a further review process of extended versions of the conference papers, three were accepted to be published in this special issue. These papers focus on analyzing or developing new intelligent methods for object recognition, image segmentation, and building combinatorial algorithms for exploring large state-space graphs. Guney et al.1 have conducted a study named “Deep Neural Network Based Toddler Tracking System (Deep-TTS),” in which a system has been developed to warn parents if their children get close to any object that might be dangerous for them. In order to monitor the children, who have just learnt to walk, the face recognition FaceNet system has been used together with an object recognition algorithm named YOLOv3. The study performs the calculation of the Euclidean distance between the toddler and the dangerous object, where YOLOv3 is used to detect the object; and the parents are warned if this distance is below the threshold value. The system's test performance is observed to be over 90% as a result of the experimental studies. In the second article, Petrovic et al.2 present an empirical methodology based on nonadversarial perturbed datasets to analyze the sensitivity of deep learning methods for ocular fundus segmentation. The authors are interested in the effect of perturbations to the input images that might happen during normal image acquisition instead of adversarial attacks. Specifically, their focus is the problem of blur: Gaussian blur, approximating an unfocused image, and motion blur, approximating either the subject or the camera moving during capture. According to their analyzes, the architectures show larger variations in sensitivity to blur, and overfitting nonessential input dataset features and resolution sensitivity are a part of the problem. The third article3 “Exploring the Blocks World State Space,” by Bădică et al. is in the area of combinatorial algorithms for exploring large state-space graphs. In this article, the authors consider the blocks world, a prototype artificial intelligence problem often used to introduce problem solving strategies using searching, planning, and reasoning. The article presents algorithms for exploring, quantifying, and visualizing the state-space graph of the blocks world. The results include: edeclarative model of the blocks world state space graph, algorithms for evaluating several metrics on this graph (number of states, average number of stacks per state, number of transitions, and average branching factor), and experimental results involving proposed algorithms. Finally, we would like to thank all authors for their contributions to this special issue by extending their papers and the reviewers for their excellent job in reviewing the articles. We also extend our thanks to Professors David W. Walker, Jinjun Chen, Nitin Auluck and Martin Berzins, editors of the Concurrency and Computation: Practice and Experience, for offering us the opportunity to prepare this special issue. We hope that this special issue will attract the attention of readers interested in innovative issues of intelligent systems. Data sharing is not applicable to this article as no new data were created or analyzed in this study.

This paper proposes a new, improved COmmon Software Measurement International Consortium function point (COSMIC FP) method that uses Artificial Neural Network (ANN) architectures based on Taguchi’s Orthogonal Array to estimate software development effort. The minimum magnitude relative error (MRE) to evaluate these architectures considering the cost effect function, the type of data used in the training, testing, and validation of the proposed models, was used. Applying the fuzzification and clustering method to obtain seven different datasets, we would like to achieve excellent reliability and accuracy of the obtained results. Besides examining the influence of four input values, we aim to reduce the risks of potential errors, increase the coverage of a wide range of different projects and increase the efficiency and success of completing many various software projects. The main contributions of our work are as follows: the influence of four input values of the COSMIC FP method on the change of mean (MRE), development of two simple ANN architectures, the attainment of a small number of performed iterations in software effort estimation (less than 7), reduced software effort estimation time, the use of different values of the International Software Benchmarking Standards Group and other datasets used in the experiment.

Accurate assessment of software project development using the proper artificial intelligence tools can be a significant challenge for success in the software industry. This paper aims to minimize the relative error in software estimation using the proposed model of an artificial neural network (ANN) based on Taguchi's orthogonal vector plan. By selecting methods of clustering and fuzzification of different project values within several used datasets such as COCOMO2000, NASA60, and Kemerer15, reducing the number and time of iterations minimizes Mean Magnitude Relative Error (MMRE) and include a wide range of observed data. Additional criteria, such as monitoring prediction, correlation, and comparison with RBF (Radial Basis Function) relative error, were used to confirm that the proposed model gives two to three times better results depending on the observed cluster. Based on the obtained results, the accuracy and reliability of the proposed model for estimating software projects were determined.

Quality of life (QoL) is one of the major issues for cancer patients. With the advent of medical databases containing large amounts of relevant QoL information it becomes possible to train predictive QoL models by machine learning (ML) techniques. However, the training of predictive QoL models poses several challenges mostly due to data privacy concerns and missing values in patient data. In this paper, we analyze several classification and regression ML models predicting QoL indicators for breast and prostate cancer patients. Two different approaches are employed for imputing missing values. The examined ML models are trained on datasets formed from two databases containing a large number of anonymized medical records of cancer patients from Sweden. Two learning scenarios are considered: centralized and federated learning. In the centralized learning scenario all patient data coming from different data sources is collected at a central location prior to model training. On the other hand, federated learning enables collective training of machine learning models without data sharing. The results of our experimental evaluation show that the predictive power of federated models is comparable to that of centrally trained models for short-term QoL predictions, whereas for long-term periods centralized models provide more accurate QoL predictions.

Human activities and behaviour in different domains are usually influenced by other people’s actions and opinion. Nowadays, it is evident that there is a growing research interest in sentiment analysis, evaluation and prediction. Content from web sources and social media is frequently used when people want to see others’ opinion about different things. Our research is focused on ML-based sentiment analysis of food services reviews data. The comparison of several regression models with regards to prediction of customer satisfaction of restaurant and food services is presented. The experimental data collected from food serving businesses located in Shanghai Lujiazui Commercial Zone includes keywords extracted from the customers’ written reviews. Additionally, the data are spatially labelled enabling to conduct separate analyses for different geographical regions. As a conclusion, the keywords extracted from the customer’s reviews were suitable for the prediction of three observed satisfaction criteria: food taste, service, and environment.

Nowadays, companies are investing in brand new software, given that fact they always need help with estimating software development, effort, costs, and the period of time needed for completing the software itself. In this paper, four different architectures of Artificial Neural Networks (ANN), as one of the most desired tools for predicting and estimating effort in software development, were used. This paper aims to determine the convergence rate of each of the proposed ANNs, when obtaining the minimum relative error, first depending on the cost effect function, then on the nature of the data on which the training, testing, and validation is performed. Magnitude relative error (MRE) is calculated based on Taguchi’s orthogonal plans for each of these four proposed ANN architectures. The fuzzification method, five different datasets, the clustering method for input values of each dataset, and prediction were used to achieve the best model for estimation. Based on performed parts of the experiment, it can be concluded that the convergence rate of each proposed architecture depends on the cost effect function and the nature of projects in different datasets. By following the prediction throughout all experimental parts, it can be further confirmed that ANN-L36 gave the best results in this proposed approach. The main advantages of this model are as follows: the number of iterations is less than 10, shortened effort estimation time thanks to convergence rate, simple architecture of each proposed ANN, large coverage of different values of actual project efficiency, and minimal MMRE. This model can also serve as an idea for the construction of a tool that would be able to reliably, efficiently and accurately estimate the effort when developing various software projects.

Improvement of prediction accuracy and early detection of the Alzheimer’s disease is becoming increasingly important for managing its impact on lives of affected patients. Many machine learning approaches have been applied to support the diagnosis and prediction of this illness. In this paper we propose an approach for improving the Alzheimer’s disease classification accuracy by using data fusion of several independent clinical datasets. Data fusion was performed twofold: 1) by enriching attributes of the base dataset with the attributes of the secondary dataset and 2) by enriching the examples set of the base dataset with the examples of the secondary dataset. In both cases the missing values (for newly added attributes and/or examples) were predicted by using linear regression for numeric and naive Bayes classifier for nominal attributes. We experimented on three data sources: on a dataset of Alzheimer’s disease-impaired patients, on a dataset of patients with vascular dementia, and on a dataset of patients who have been affected by a stroke. We fused these datasets with different data fusion approaches and analysed the improvement in classification accuracy as well as the quality of the fused attributes. The experiments indicated that we obtained an increase of classification accuracy on the fused dataset compared with the accuracy obtained from individual dataset.

The problem of time-series classification witnessed the application of many techniques for data mining and machine learning, including neural networks, support vector machines, and Bayesian approaches. Somewhat surprisingly, the simple 1-nearest neighbor (1NN) classifier, in combination with the Dynamic Time Warping (DTW) distance measure, is still competitive and not rarely superior to more advanced classification methods, which includes the majority-voting k-nearest neighbor (kNN) classifier. In this paper we focus on the kNN classifier combined with the inverse-squared weighting scheme, and its interaction with constrained DTW distance. By performing experiments on the entire UCR Time Series Classification Archive we show that with proper selection of the constraint parameter r and neighborhood size k, inverse-square weighted kNN consistently outperforms 1NN.

The complexity of IoT systems and tasks that are put before them require shifts in the way resources and service provisioning are managed. The concept of edge computing is introduced to enhance IoT systems' scalability, reactivity, efficiency, and privacy. In this paper, we present an edge computing solution for resource management of context-aware decision-making processes distributed between IoT gateways. The solution performs decision-making process management for smart actuation, based on analysis of sensory data streams, and context-informed edge computing resource and service provisioning management based on topology and operational changes. Our architectural solution showcases the first version of a Resource Management Framework - a generic framework for software resource orchestration best-suited to IoT platforms with event-driven, publish-subscribe communication mechanisms. Proof of concept experiments that are executed in a simulated edge computing testbed validate our solution's performance in improving the resilience and responsiveness of the edge computing system when there are operational and topology changes. Furthermore, the framework addresses the recovery of failed decision-making processes, impacting the overall health of the underlying IoT system.

[No abstract available]

We have recently proposed a mathematical model of collective profitability, in semi-competitive intermediation networks. In this work, we are interested in determining optimal pricing strategies of network participants. The optimization criterion is defined using the Nash social welfare function. We provide theoretical results of existence of such strategies, as well as computational experimental results, based on nonlinear convex mathematical optimization.

In the domain of time-series classification, one simple but persistently successful method is the 1-nearest neighbour (1NN) classifier coupled with an elastic distance measure such as Dynamic Time Warping (DTW). In this paper we evaluate the performance of DTW when constrained using the Itakura parallelogram, and compare it with the more commonly used Sakoe-Chiba band, as well as with the unconstrained DTW. Results show that although the Itakura parallelogram is generally inferior to the Sakoe-Chiba band, it is still superior to unconstrained DTW. Furthermore, on individual data sets the Itakura parallelogram can produce superior results, warranting further investigation into the merits of its use with DTW and other elastic distance measures for time-series classification.

The overarching connectivity of "things" in the Internet of Things presents an appealing environment for innovation and business ventures, but also brings a certain set of security challenges. Engineering secure Internet of Things systems requires addressing the peculiar circumstances under which they operate: constraints due to limited resources, high node churn, decentralized decision making, direct interfacing with end users etc. Thus, techniques and methodologies for building secure and robust Internet of Things systems should support these conditions. In this paper, we are presenting a description of the CAAVI-RICS framework, a novel security review methodology tightly coupled with distributed, Internet of Things and fog computing systems. With CAAVI-RICS we are exploring credibility, authentication, authorization, verification, and integrity (CAAVI) through explaining the rationale, influence, concerns and security solutions (RICS) that accompany them. Our contribution is a thorough systematic categorization and rationalization of security issues, covering the security landscape of Internet of Things/fog computing systems, as well as contributing to the discussion on the aspects of fog computing security and state-of-the-art solutions. Specifically, in this paper we explore the Authentication in Internet of Things systems through the RICS review methodology.

This paper presents different approaches, experiences and perspectives of using technologies in higher education institutions. Particular case studies of application of social media (especially wikis), game-based learning and various technology-enhanced learning tools in different courses at several Serbian institutions are presented. In-house developed intelligent tutoring system Protus and possibilities to enhance it by software agents and eye-tracking are also shown in detail. Our experiences of using different technology-enhanced learning tools and mechanisms showed that educational processes must be modernized and enhanced by technological progress.

Personalization of the e-learning systems according to the learner’s needs and knowledge level presents the key element in a learning process. E-learning systems with personalized recommendations should adapt the learning experience according to the goals of the individual learner. Aiming to facilitate personalization of a learning content, various kinds of techniques can be applied. Collaborative and social tagging techniques could be useful for enhancing recommendation of learning resources. In this paper, we analyze the suitability of different techniques for applying tag-based recommendations in e-learning environments. The most appropriate model ranking, based on tensor factorization technique, has been modified to gain the most efficient recommendation results. We propose reducing tag space with clustering technique based on learning style model, in order to improve execution time and decrease memory requirements, while preserving the quality of the recommendations. Such reduced model for providing tag-based recommendations has been used and evaluated in a programming tutoring system.

Abstract We propose a new framework based on Belief‐Desire‐Intention multi‐agent systems for the macroscopic modelling and simulation of continuous dynamic systems. The main idea is to break down the target system model into a collection of autonomous and loosely coupled interacting components endowed with clean message‐based interfaces and local intelligence. Each component is then mapped to a Belief‐Desire‐Intention agent that captures its state as a set of logical facts and its behavioural patterns as a set of plans. The system model can be described as a multi‐agent programme that is specified using the state‐of‐the‐art Jason agent‐oriented programming language. The approach is evaluated by considering a generalized graph‐based model of predator–prey systems. Our approach supports the configuration of the multi‐agent model with various differential equations integration methods, as required by the specific problem.

Current Research Information Systems (CRISs) offer great opportunities for assessments of institutional research outputs and extraction of useful and actionable knowledge based on various data-analysis techniques. However, many of these opportunities have not been explored in depth, especially in c ulture-sensitive areas such as gender-based analysis of research productivity and collaboration. In this paper we present GERBER, a network-based methodology and accompanying tool for gender-based analysis of publication data stored in institutional CRISs. GERBER relies on statistically robust techniques applied on weighted co-authorship networks whose nodes are enriched with different types of researcher evaluation metrics. The functionality of GERBER is demonstrated on publication data stored in the institutional CRIS of the Faculty of Sciences, University of Novi Sad, Serbia. The obtained results show that GERBER enables institutional research managers and policy makers to detect gender inequalities and homophily in research productivity and collaboration. Finally, we discuss different possibilities to integrate GERBER with CRISs in order to facilitate continuous gender-based evaluation of researchers.

The purpose of the research is to provide an empirical insight into the gameplay preferences and multiple intelligences profile of early adolescents living in rural areas. It was found that the early adolescent males living in rural areas played digital games about two times more than females did, and that the average weekly gameplay time increased as students grew. Females achieved significantly higher levels of Bodily/kinesthetic intelligence than males did. Gender and age can be observed as predictors of average daily gameplay time. The levels of Visual/spatial, Verbal/linguistic, Interpersonal and Natural intelligence can be observed as the predictors of average weekly gameplay time. Socio-cultural influence of the living environment presented a significant factor for digital gameplay habits and multiple intelligence profile of early adolescents.

Conceptual modelling education is often a part of any Informatics and Computer Science curricula but is also included in other study programs where conceptual modelling knowledge is needed. Conceptual modelling is well defined and numerous sources are available in existing literature. The steps to develop conceptual models are easy to learn and support is possible also via different tools, based on a theoretical background, mostly through the Peter Chen approach. In general, such an educational approach can be described as a standard conceptual modelling education. But, unfortunately, standard conceptual modeling education does not inform culturally-aware developers of conceptual models. The confidence and assurance that cross cultural issues and cultural awareness will be considered in the phase of conceptual modelling could be reached on one side by upgrading the conceptual modeling approach with a cultural point of view, as well as by introducing cultural issues into conceptual modelling education. Students need to be aware of basic cultural issues and concepts and their influence on conceptual modeling. In the long term, the mentioned content should become part of standard conceptual modeling education. This means changes in the content of curricula as well as a change in thinking by students and teachers.

Outlier detection in high-dimensional data presents various challenges resulting from the “curse of dimensionality.” A prevailing view is that distance concentration, i.e., the tendency of distances in high-dimensional data to become indiscernible, hinders the detection of outliers by making distance-based methods label all points as almost equally good outliers. In this paper, we provide evidence supporting the opinion that such a view is too simple, by demonstrating that distance-based methods can produce more contrasting outlier scores in high-dimensional settings. Furthermore, we show that high dimensionality can have a different impact, by reexamining the notion of reverse nearest neighbors in the unsupervised outlier-detection context. Namely, it was recently observed that the distribution of points' reverse-neighbor counts becomes skewed in high dimensions, resulting in the phenomenon known as hubness. We provide insight into how some points (antihubs) appear very infrequently in k-NN lists of other points, and explain the connection between antihubs, outliers, and existing unsupervised outlier-detection methods. By evaluating the classic k-NN method, the angle-based technique designed for high-dimensional data, the density-based local outlier factor and influenced outlierness methods, and antihub-based methods on various synthetic and real-world data sets, we offer novel insight into the usefulness of reverse neighbor counts in unsupervised outlier detection.

Developing teaching materials is a time-consuming and expensive activity. Thus, over the years, several consortia have created joint materials to benefit from them. It is expected that shared teaching materials are a means to save effort in its development, to transfer methodological and technical knowledge between different university staff, and to exchange experience in practical application. However, does it really pay off considering the diversity of different educational environments and the difficulties of using externally produced materials, rather than dedicated individual ones? This paper reports on the experience gained in a multi-country project. Both success factors and problems are outlined. It turned out that sharing an educational repository by several partners supports maintaining quality of teaching materials by joint efforts and distributed contributions.

This paper outlines the primary steps to investigate if artificial agents can be considered as true substitutes of humans. Based on a Socially augmented microworld (SAM) human tracking behavior was analyzed using time series. SAM involves a team of navigators jointly steering a driving object along different virtual tracks containing obstacles and forks. Speed and deviances from track are logged, producing high-resolution time series of individual (training) and cooperative tracking behavior. In the current study 52 time series of individual tracking behavior on training tracks were clustered according to different similarity measures. Resulting clusters were used to predict cooperative tracking behavior in fork situations. Results showed that prediction was well for tracking behavior shown at the first and, moderately well at the third fork of the cooperative track: navigators switched from their trained to a different tracking style and then back to their trained behavior. This matches with earlier identified navigator types, which were identified on visual examination. Our findings on navigator types will serve as a basis for the development of artificial agents, which can be compared later to behavior of human navigators.

High-dimensional data arise naturally in many domains, and have regularly presented a great challenge for traditional data mining techniques, both in terms of effectiveness and efficiency. Clustering becomes difficult due to the increasing sparsity of such data, as well as the increasing difficulty in distinguishing distances between data points. In this paper, we take a novel perspective on the problem of clustering high-dimensional data. Instead of attempting to avoid the curse of dimensionality by observing a lower dimensional feature subspace, we embrace dimensionality by taking advantage of inherently high-dimensional phenomena. More specifically, we show that hubness, i.e., the tendency of high-dimensional data to contain points (hubs) that frequently occur in k-nearest-neighbor lists of other points, can be successfully exploited in clustering. We validate our hypothesis by demonstrating that hubness is a good measure of point centrality within a high-dimensional data cluster, and by proposing several hubness-based clustering algorithms, showing that major hubs can be used effectively as cluster prototypes or as guides during the search for centroid-based cluster configurations. Experimental results demonstrate good performance of our algorithms in multiple settings, particularly in the presence of large quantities of noise. The proposed methods are tailored mostly for detecting approximately hyperspherical clusters and need to be extended to properly handle clusters of arbitrary shapes.

Effective team formation and the influence of team structure on the results students accomplish on team assignments are discussed in this paper. Also some strategies for team formation and a particular software tool have been developed to help automate that process.

A course on “Software Project Management” created as a part of DAAD project was conducted for the seventh time at the Department of Mathematics and Informatics in Novi Sad during the school-year 2010/11. At the beginning, it has been a part of the “Business Informatics” curriculum, as an elective course for the students of the final semester of studies. After realizing its importance, the course was moved, and became an obligatory subject of the first year of master studies of “Software Engineering” direction. It has been realized completely in accordance with Bologna declaration and continual assessment. Course includes practical work on a real world project, and class hours devoted to discussions and role-playing games considering the situations of programmer's real life. Using funds of another project, Austrian WUS association program, respectable literature on a subject was purchased, and an appropriate textbook was written. In this paper, we will present our long-lasting experiences with the development and teaching of this course, and also results of a survey covering students' opinions about the course.

Collaborative tagging systems have grown in popularity over the Web in the last years based on their simplicity to categorize and retrieve content using open-ended tags. Besides helping user to organize his/her personal collections, a tag also can be regarded as a user's or expert's personal opinion expression. Thus, the tagging information can be used to make recommendations. In this paper, an innovative architecture for a tag-based recommender system dedicated to the e-learning environments is introduced. This system could support learners by recommending tags and learning resources, online learning activities or optimal browsing pathways, based on their preferences, learning style, knowledge level and the browsing history of other learners with similar characteristics.

Most machine-learning tasks, including classification, involve dealing with high-dimensional data. It was recently shown that the phenomenon of hubness, inherent to high-dimensional data, can be exploited to improve methods based on nearest neighbors (NNs). Hubness refers to the emergence of points (hubs) that appear among the k NNs of many other points in the data, and constitute influential points for kNN classification. In this paper, we present a new probabilistic approach to kNN classification, naive hubness Bayesian k-nearest neighbor (NHBNN), which employs hubness for computing class likelihood estimates. Experiments show that NHBNN compares favorably to different variants of the kNN classifier, including probabilistic kNN (PNN) which is often used as an underlying probabilistic framework for NN classification, signifying that NHBNN is a promising alternative framework for developing probabilistic NN algorithms.

In the context of many data mining tasks, high dimensionality was shown to be able to pose significant problems, commonly referred to as different aspects of the curse of dimensionality. In this paper, we investigate in the time-series domain one aspect of the dimensionality curse called hubness, which refers to the tendency of some instances in a data set to become hubs by being included in unexpectedly many k-nearest neighbor lists of other instances. Through empirical measurements on a large collection of time-series data sets we demonstrate that the hubness phenomenon is caused by high intrinsic dimensionality of time-series data, and shed light on the mechanism through which hubs emerge, focusing on the popular and successful dynamic time warping (DTW) distance. Also, the interaction between hubness and the information provided by class labels is investigated, by considering label matches and mismatches between neighboring time series. Following our findings we formulate a framework for categorizing time-series data sets based on measurements that reflect hubness and the diversity of class labels among nearest neighbors. The framework allows one to assess whether hubness can be successfully used to improve the performance of k-NN classification. Finally, the merits of the framework are demonstrated through experimental evaluation of 1-NN and k-NN classifiers, including a proposed weighting scheme that is designed to make use of hubness information. Our experimental results show that the examined framework, in the majority of cases, is able to correctly reflect the circumstances in which hubness information can effectively be employed in k-NN time-series classification.

The vector space model (VSM) is a popular and widely applied model in information retrieval (IR). VSM creates vector spaces whose dimensionality is usually high (e.g., tens of thousands of terms). This may cause various problems, such as susceptibility to noise and difficulty in capturing the underlying semantic structure, which are commonly recognized as different aspects of the "curse of dimensionality." In this paper, we investigate a novel aspect of the dimensionality curse, which is referred to as hubness and manifested by the tendency of some documents (called hubs) to be included in unexpectedly many search result lists. Hubness may impact VSM considerably since hubs can become obstinate results, irrelevant to a large number of queries, thus harming the performance of an IR system and the experience of its users. We analyze the origins of hubness, showing it is primarily a consequence of high (intrinsic) dimensionality of data, and not a result of other factors such as sparsity and skewness of the distribution of term frequencies. We describe the mechanisms through which hubness emerges by exploring the behavior of similarity measures in high-dimensional vector spaces. Our consideration begins with the classical VSM (tf-idf term weighting and cosine similarity), but the conclusions generalize to more advanced variations, such as Okapi BM25. Moreover, we explain why hubness may not be easily mitigated by dimensionality reduction, and propose a similarity adjustment scheme that takes into account the existence of hubs. Experimental results over real data indicate that significant improvement can be obtained through consideration of hubness.

High dimensionality can pose severe difficulties, widely recognized as different aspects of the curse of dimensionality. In this paper we study a new aspect of the curse pertaining to the distribution of k-occurrences, i.e., the number of times a point appears among the k nearest neighbors of other points in a data set. We show that, as dimensionality increases, this distribution becomes considerably skewed and hub points emerge (points with very high k-occurrences). We examine the origin of this phenomenon, showing that it is an inherent property of high-dimensional vector space, and explore its influence on applications based on measuring distances in vector spaces, notably classification, clustering, and information retrieval.

A crucial operation in memory-based collaborative filtering (CF) is determining nearest neighbors (NNs) of users/items. This paper addresses two phenomena that emerge when CF algorithms perform NN search in high-dimensional spaces that are typical in CF applications. The first is similarity concentration and the second is the appearance of hubs (i.e. points which appear in $k$-NN lists of many other points). Through theoretical analysis and experimental evaluation we show that these phenomena are inherent properties of high-dimensional space, unrelated to other data properties like sparsity, and that they can impact CF algorithms by questioning the meaning and representativeness of discovered NNs. Moreover, we show that it is not easy to mitigate the phenomena using dimensionality reduction. Studying these phenomena aims to provide a better understanding of the limitations of memory-based CF and motivate the development of new algorithms that would overcome them.

Abstract Case‐based reasoning (CBR) is the area of artificial intelligence where problems are solved by adapting solutions that worked for similar problems from the past. This technique can be applied in different domains and with different problem representations. In this paper, a system curve base generator (CuBaGe) is presented. This framework is designed to be a domain‐independent prediction system for the analysis and prediction of curves and time‐series trends, based on the CBR technology.CuBaGeemploys a novel curve representation method based on splines and a corresponding similarity function based on definite integrals. This combination of curve representation and similarity measure showed excellent results with sparse and non‐equidistant time series, which is demonstrated through a set of experiments. Copyright © 2008 John Wiley & Sons, Ltd.

Case-based reasoning (CBR) is the area of the artificial intelligence in which the new problems are solved by adapting the solutions of the previously successfully solved similar problems. Medicine is a suitable domain for application of CBR because the knowledge of medical experts consists of mixture of textbook knowledge and experience, which consists of cases. Architecture of the system for the diagnoses of multiple sclerosis disease, based on CBR is proposed in this paper.