Search

The paper proposes EXPLODE, a new model of exploratory learning environment for teaching and learning neural networks. The EXPLODE model is about pedagogically instrumenting a software development environment to transform it into an exploratory learning environment for neural networks. Such an environment is particularly aimed for students who are skilled in programming and meets typical challenges in teaching and learning neural networks, related to the lack of prerequisite knowledge in mathematics. By providing such students with a familiar learning environment and allowing them to programmatically experiment with neural networks, the EXPLODE model aims at improving the students' learning outcomes and learning experience. The effectiveness of the model was evaluated in an experimental study with 77 final-year IT students. The results have shown that the students from the experimental group, that is, those who were exposed to the EXPLODE model, scored higher on the knowledge test. Furthermore, the perceived learning experience of the experimental group was better than that of the control group. The results have also suggested that the proposed model helps students better understand learning topics that require practical experience and facilitates a deeper understanding of the internal operations of neural networks.

The paper presents ideas and initiatives from two ongoing Erasmus+ projects funded by the European Commission. Both projects use e-Learning as an enabler for communicating interesting and important learning contents that are believed to increase and improve employability prospects for the targeted groups of learners. The WINnovators project targets young women with reduced opportunities for quality education in different areas of STEM/STEAM and intends to demonstrate how their entrepreneurial skills can gradually grow by providing them with e-Learning incentives in such areas. The YNSPEED project intends to offer free MOOC courses addressing hot topics of modern society – artificial intelligence, sustainable development, and fake news. In the YNSPEED project, the target group of learners are young people (aged 16-29). It is believed that their interest in such important topics can be boosted by communicating the corresponding relevant learning content in a carefully designed way. The Shine a light metaphor is a common point and a common approach in both projects – e-Learning technology is used to facilitate informal education for the targeted learners and to indicate directions that often get either omitted or misunderstood in traditional schools.

The Gaussian Conditional Random Fields (GCRF) algorithm and its extensions are used for machine learning regression problems in which the attributes of objects and the correlation between objects should be considered when making predictions. These algorithms can be applied in different domains where problems can be seen as graphs, but their implementation requires complex calculations and good programming skills. This paper presents an open source software package that includes a tool with graphical user interface (GCRFs tool) and Java library (GCRFs library). GCRFs tool is software that integrates various GCRF-based algorithms and supports training and testing of those algorithms on real-world datasets. The main goal of GCRFs tool is to provide a straightforward and user-friendly graphical user interface that will simplify the use of GCRF-based algorithms. GCRFs Java library contains basic classes for GCRF concepts and can be used by researchers who have experience in Java programming. Also, this paper presents the results of a pilot usability evaluation of the GCRFs tool, where the software was evaluated with expert and non-expert users. This evaluation gave us detailed insight into the experiences and opinions of the users and helped us outline priorities for future development.

Background: Technology-enhanced assessment (TEA) is a broad term that encompasses the diverse methods, by which technology can be used to support the management and delivery of assessment in educational institutions, in workplace and lifelong learning, in professional training and development, and so on. Purpose of Study: There is a new initiative at the University of Belgrade, Serbia, to explore computing technology underpinnings of the practice of TEA (computing underpinnings of the practice [CUP] of TEA), at universities and in schools in the country. Sources of Evidence: The CUP of TEA initiative starts from the current insight into current practices of assessment processes in different educational institutions. The data collected include assessment process descriptions from these institutions, specification of technologies used for the purpose, and specification of the educational profiles of participants in the process. Main Argument: The CUP of TEA initiative develops: A novel and comprehensive framework for TEA to lay the foundation for systematic, accurate, effective, and innovative application and use of TEA in educational institutions; specifically designated sets of information and communication technology (ICT) tools and services – ICT toolsets – that support practical application of TEA; several application cases that implement the TEA framework and use different ICT toolsets for TEA in different educational contexts, showcasing the integration of best TEA practices into curricula; and a set of policies that enable educational institutions and their teachers to officially establish the use of the framework for TEA in their courses and daily work. The CUP of TEA initiative covers both formative and summative assessment and supports formal, non-formal, and informal learning and assessment. Conclusions: The ultimate goal of the initiative is not just better, faster, and more informative assessment. It is the creation of a robust teaching and learning asse

There has been a lot of research related to the prediction of students’ interest in a course and course performance. The findings are especially important for those fields (and courses) with high failure and dropout rates, such as computer science. Although many research in computer science education involve various motivational and learning strategy frameworks (like achievement goal orientation and Motivated Strategies for Learning Questionnaire – MSLQ), very few involve interest, attendance and self-evaluation. In this study, the aim was to see whether students’ achievement goal orientation and self- evaluation of their pre-faculty programming knowledge are related to course performance, attendance and interest in an introductory computer science course. Additionally, we wanted to see if attendance and interest relate to course performance. The findings suggest that only task-approach has a positive correlation with final test scores (but not with mid-term test scores), and that all AGO constructs except other-avoidance correlate positively with interest. Also, attendance has a positive correlation with mid-term test scores (but not with final test scores) while interest has a positive correlation with both mid-term and final test scores. Finally, we suspect that attendance and mid-term test findings may be somewhat influenced by course and faculty policies.

Artificial Intelligence (Al) has become one of the mostfrequently used terms in the technical jargon (and often in not-so-technical jargon). Recent advancements in the field ofAI have certainly contributed to the AI hype, and so have numerous applications and results of using AI technology in practice. Still, just like with any other hype, the AI hype has its controversies. This paper critically examines developments in the field ofAI from multiple perspectives research, technological, social and pragmatic. Part of the controversies of the AI hype stem from the fact that people use the term AI differently, often without a deep understanding of the wider context in which AI as a field has been developing since its inception in Mid 1950s.

Samoregulisano učenje predstavlja formu učenja koja stavlja akcenat na samog učenika i pruža mu mogućnost da upravlja i kontinuirano reguliše svoj proces učenja. Poznato je da društveni kontekst i interakcije koje se ostvaruju tokom učenja imaju uticaj na akcije učenika i da se samoregulacija može posmatrati kao regulacija individue u okviru socijalnog okruženja. Međutim, postojeći sistemi za učenje na Vebu pružaju učenicima ograničene mogućnosti za samoregulaciju tokom učenja u neformalnim i informalnim kontekstima. Predmet ove doktorske disertacije je dizajniranje i projektovanje softverskog sistema koji je namenjen samoregulisanom učenju na Vebu. Cilj je da ovaj softverski sistem poseduje funkcionalnosti koje podržavaju procese samoregulisanog učenja, da omogućavaju uređivanje kohorti, kurseva, materijala i aktivnosti učenja, kao i da omogućavaju kolektivnu regulaciju između učenika. Prototip softverskog sistema, koji je razvijen na osnovu predloženog dizajna i arhitekture, evaluiran je u okviru MOOC kursa u trajanju od 9 nedelja. Dobijeni rezultati su pokazali da postoje izražene asocijacije između funkcionalnosti zasnovanim na socijalnim interakcijama i procesa samoregulisanog učenja koji je zasnovan na kolektivnoj regulaciji. Konkretno, pokazano je da postoje asocijacije između funkcionalnosti koje omogućavaju uvid u aktivnosti i progres drugih učenika sa procesima samoregulisanog učenja koji se odnose na izvršenje zadataka učenika. Takođe, pronađene su razlike u ovim asocijacijama kod učenika sa različitim demografskim i motivacionim karakteristikama., Self-regulated learning is a form of learning that puts emphasis on a learner and gives them the ability to manage and continually regulate their learning process. It is known that the social context and interactions that occur during learning can impact students' actions and that self-regulation can be viewed as a regulation of an individual within their social context. However, existing web learning systems provide students with limited functionalities to self-regulate their learning in informal and informal contexts. The topic of this doctoral dissertation is the design and architecture of a software system that enables self-regulated learning on the Web. The goal of this system is to enable a learner with functionalities that support self-regulated learning processes, to enable management of cohorts, courses, materials and learning activities, as well as to allow collective regulation among learners. A software prototype, developed based on the proposed design and architecture, was evaluated within the MOOC course for 9 weeks. The obtained results showed that there are significant associations between functionalities based on social interactions and the self-regulated learning process based on collective regulation. In particular, it has been shown that there are associations between functionalities that allow insight into the learning activities and progress of other students with self-regulated learning processes related to engagement in learning. Also, variations in these associations were found among students with different demographic and motivational characteristics.

Otvoreni bedževi su obećavajuća edukativna tehnologija za priznavanje postignuća, veštine, kvaliteta ili interesa pojedinca ili grupe. Poslednjih godina, pojavio se veći broj softverskih sistema zasnovanih na Infrastrukturi Otvorenih bedževa (eng. Open Badges Infrastructure - OBI) koji podržavaju različite prakse nastave/učenja i izdavanja bedževa. Međutim, njihova upotrebljivost u konkretnom edukativnom okruženju može značajno da varira. To nastavnicima donosi velike izazove prilikom odabira jednog ovakvog sistema za konkretnu edukativnu primenu. Cilj ovog istraživanja je da predloži potencijalne metrike upotrebljivosti - opšte i tehnološki-specifične za sisteme zasnovane na OBI specifikaciji, koje se mogu koristiti efektivno prilikom odabira ove edukativne tehnologije. Glavni rezultat istraživanja je metodološki okvir koji je obuhvatio standardne opšte metrike upotrebljivosti (npr., ocena opažene upotrebljivosti na osnovu Skale Upotrebljivosti Sistema) i određene tehnološki-specifične metrike upotrebljivosti koje su razvijene u okviru istraživanja (npr., ocena lakoće deljenja bedževa, ocena zadovoljstva podrškom za kreiranje bedževa, itd.). Predloženi pristup je ispitan u empirijskoj studiji zasnovanoj na mešovitom metodu istraživanja. Rezultati studije su obezbedili dokaze o efektivnosti pristupa u datom kontekstu. Pokazalo se da su predložene kategorije metrika komplementarne i da su metrike diferencirale evaluirane sisteme na zadovoljavajući način. Uz to, rezultati kvalitativne evaluacije su otkrili glavne probleme upotrebljivosti evaluiranih sistema. Pristup se može primeniti u različitim edukativnim okruženjima s obzirom na to da su predložene metrike relativno jednostavne za prikupljanje i interpretaciju., Open badges are a promising educational technology for recognizing the achievements, skills, qualities and interests of an individual or а group. In recent years, there has been a growing number of software systems based on the Open Badges Infrastructure (OBI) which support different teaching/learning and badging practices. However, their usability in a specific educational environment can vary significantly. This brings some great challenges for teachers when choosing one such system for a concrete educational application. The aim of this research is to propose some prospective usability metrics - general and technology-specific for OBI-compliant systems, which can be used effectively when selecting this educational technology. The main research result is a methodological framework that has included some standard general usability metrics (e.g., score of perceived usability based on System Usability Scale) and the technology-specific usability metrics that have been developed within the research (e.g., score of ‘ease of sharing badges’, score of ‘support for badge creation satisfaction’, etc.). The proposed approach has been examined in an empirical study based on a mixed research method. The results of the study provided evidence of the effectiveness of the proposed approach in the given context. It has been shown that the proposed categories of metrics are complementary and that the metrics have differentiated the evaluated systems in a satisfactory manner. In addition, the results of qualitative evaluation revealed the main usability problems of the evaluated systems. The approach can be applied in different educational settings, given that the proposed metrics are relatively simple to collect and interpret.

In order to be effective, programmers need to master both theory and practical skills. The key for acquiring good programming skills is the active work: probing code examples, solving problems, doing assignments, developing programs, etc. Unfortunately, many educational materials do not promote active learning: the code must first be retyped or copied before it can be executed, the solution cannot be validated without manual testing or the involvement of teachers and colleagues, the materials cannot be easily distributed and updated, etc. The approach presented in this paper advocates using executable Java and R projects organized as Git code repositories (hosted on GitHub) as means of encouraging active learning. Executable solutions are provided within each project and, in the case of Java assignments, automated JUnit tests are also present so students can receive immediate feedback on the correctness of their solution. The approach had been implemented and evaluated in two programming courses, and the results suggest that students have a very positive attitude towards the use of such materials. The teachers’ experiences show that, once implemented, the approach offers many benefits in terms of collaborative making of materials, their distribution and maintenance.

Well-developed programming (technical) skills are very important for software engineers, information systems engineers and programmers in general. However, they must also possess relevant personal skills (soft skills) to be successful at the workplace (eg, collaboration, solving real-world problems and communication). The latter, however, are rarely assessed and acknowledged in regular software engineering courses. This paper describes the results of a small case study involving an extracurricular Java programming course in which, in addition to knowledge and skills in relevant technologies, students' soft skills were also assessed. As part of the assessment, students have been awarded Open Badges. The study was exploratory in nature, aimed at examining Open Badges as a motivational mechanism, students' engagement in attaining soft skills and students' perception of soft skills and Open Badges. The results suggest that Open Badges may not be so effective in motivating students to complete the assignments nor attend the course, although students' perception of Open Badges is generally positive. Soft skills were generally perceived as important as hard skills. Students' engagement in attaining soft skills could be affected by assignment announcement time and its level of difficulty.

U sklopu doktorske disertacije izvršeno je istraživanje vezano za automatsko raspoređivanje elemenata dijagrama. Kroz analizu postojećih rešenja uočen je prostor za poboljšanja, posebno po pitanju raznovrsnosti dostupnih algoritama i pomoći korisniku pri izboru najpogodnijeg od njih. U okviru istraživanja proučavan, implementiran i u pojedinim slučajevima unapređen je širok spektar algoritama za crtanje i analizu grafova. Definisan je postupak automatskog izbora odgovarajućeg algoritma za raspoređivanje elemenata grafova na osnovu njihovih osobina. Dodatno, osmišljen je jezik specifičan za domen koji korisnicima grafičkih editora pruža pomoć u izboru algoritma za raspoređivanje, a programerima brže pisanje koda za poziv željenog algoritma., This thesis presents a research aimed towards the problem of automatically laying out elements of a diagram. The analysis of existing solutions showed that there is some room for improvement, especially regarding variety of available algorithms. Also, none of the solutions offer possibility of automatically choosing an appropriate graph layout algorithm. Within the research, a large number of different algorithms for graph drawing and analysis were studied, implemented, and, in some cases, enhanced. A method for automatically choosing the best available layout algorithm based on properties of a graph was defined. Additionally, a domain-specific language for specifying a graph’s layout was designed.

Овом докторском дисертацијом предложен је систем за састављање докумената у правосуђу базиран на знању. Састављање судских писмена је препознато као један од изазова с којим се сусрећу правници на почетку своје каријере у чему им помоћ најчешће пружају старије колеге, док је правно неуким странкама у састављању поднесака најчешће неопходна помоћ искусних правника. За представљање знања на којем се базира вештина састављања ових докумената коришћена су два облика правног знања, експлицитно које је садржано у правним нормама и имплицитно које се стиче искуством у састављању правних докумената. Предложени су модели машински читљивог и разумљивог формата правних норми, машински читљивог и разумљивог формата докумената у правосуђу и систем за састављање докумената у правосуђу. Систем је погодан за коришћење у образовању правника јер су кориснику објашњене везе између унетих чињеница и навода у генерисаном документу. Документи сачињени на овај начин су машински читљиви и разумљиви чиме се доприноси квалитету отворених података при њиховом објављивању. У циљу повећања транспарентности правосуђа предложене су и смернице за превазилажење најчешћих недостатака отворених судских података. Такође, предложен систем представља један корак ка аутоматизацији рада судске писарнице., Ovom doktorskom disertacijom predložen je sistem za sastavljanje dokumenata u pravosuđu baziran na znanju. Sastavljanje sudskih pismena je prepoznato kao jedan od izazova s kojim se susreću pravnici na početku svoje karijere u čemu im pomoć najčešće pružaju starije kolege, dok je pravno neukim strankama u sastavljanju podnesaka najčešće neophodna pomoć iskusnih pravnika. Za predstavljanje znanja na kojem se bazira veština sastavljanja ovih dokumenata korišćena su dva oblika pravnog znanja, eksplicitno koje je sadržano u pravnim normama i implicitno koje se stiče iskustvom u sastavljanju pravnih dokumenata. Predloženi su modeli mašinski čitljivog i razumljivog formata pravnih normi, mašinski čitljivog i razumljivog formata dokumenata u pravosuđu i sistem za sastavljanje dokumenata u pravosuđu. Sistem je pogodan za korišćenje u obrazovanju pravnika jer su korisniku objašnjene veze između unetih činjenica i navoda u generisanom dokumentu. Dokumenti sačinjeni na ovaj način su mašinski čitljivi i razumljivi čime se doprinosi kvalitetu otvorenih podataka pri njihovom objavljivanju. U cilju povećanja transparentnosti pravosuđa predložene su i smernice za prevazilaženje najčešćih nedostataka otvorenih sudskih podataka. Takođe, predložen sistem predstavlja jedan korak ka automatizaciji rada sudske pisarnice., This thesis proposes a knowledge-based judicial document assembly system. Document assembly is recognized as one of the issues facing junior lawyers at the beginning of their professional career causing them to rely on their senior colleague's experience. Also, filings’ preparation is a challenging task for nonlayers. Therefore they usually need to hire a lawyer. Knowledge required for document assembly can be divided into stated knowledge found in regulations and legal textbooks and tacit knowledge gained through experience. This thesis introduces a machine-readable and machine-understandable format of legal norms, a machine-readable and machine-understandable format of judicial documents, and a system for judicial document assembly. The assembly system has its potential in the education of law students providing them explanations how fragments of generated document relates to case facts. Also, this approach improves quality of open judicial data and increases transparency of judiciary because generated documents are machine-readable and machineunderstandable by default. In addition, a set of guidelines for opening judicial data is proposed. Finally, a machine-readable and machineunderstandable format of generated documents is a step toward automatic document processing at court clerk's office.

Predmet istraživanja ovog rada su modeli strukturne regresije, koji su dizajnirani da koriste veze između objekata prilikom predviđanja izlaznih vrijednosti. Drugim riječima, modeli strukturne regresije razmatraju atribute objekata i veze između objekata kako bi dali što tačnije predviđanje. Gaussian Conditional Random Fields (GCRF) model je jedan od najčešće korišćenih modela strukturne regresije koji integriše predikciju tradicionalnih modela nadgledanog učenja (nestrukturnih prediktora) i vezu između objekata u cilju tačnije predikcije. Glavna pretpostavka ovog modela je da su dva objekta koja su usko povezana veoma slični jedan drugom i samim tim vrijednosti njihovih izlaznih varijabli treba da budu slični. Sličnost između objekata u GCRF modelu mora da bude simetrična, ali u velikom broju realnih primjera objekti su nesimetrično povezani. U radu je predstavljeno proširenje GCRF modela koje uzima u obzir asimetričnu sličnost između objekata (nazvan usmjereni GCRF - Directed GCRF). Na sintetičkim i realnim setovima podataka pokazano je da novi model daje tačnije predviđanje od standardnog GCRF modela i tradicionalnih nestrukturnih prediktora. Rad obuhvata i razvoj softverskog alata otvorenog koda koji integriše različite vrste GCRF modela i omogućava treniranje i testiranje tih modela na različitim setovima podataka, preko grafičkog korisničkog interfejsa. Izvršena je evaluacija alata sa korisnicima različitih profila i različitog znanja iz oblasti mašinskog učenja. Rezultati su potvrdili da je alat je intuitivan i lak za korišćenje kako za eksperte, tako i za početnike i istraživače iz različitih domena kojima GCRF model može pomoći da dođu do željenih informacija., The subject of this dissertation are structured regression models that are designed to use relationships between objects for predicting output variables. In other words, structured regression models consider the attributes of objects and dependencies between objects to make predictions as accurately as possible. Gaussian Conditional Random Fields (GCRF) model is commonly used structured regression model that incorporates the outputs of traditional supervised learning models (unstructured predictors) and the correlation between output variables in order to achieve a higher prediction accuracy. A main assumption in the GCRF model is that if two objects are closely related, they should be more similar to each other and they should have similar values of the output variable. The similarity considered in GCRF is symmetric. However, in many real-world examples objects are asymmetrically linked. This dissertation presents extension of GCRF model that considers asymmetric similarities between objects (called Directed GCRF). The effectiveness of new model is characterized on synthetic datasets and real-world datasets, on which it was more accurate than the standard GCRF model and baseline unstructured predictors. This dissertation also presents development of an open-source software tool that integrates various GCRF methods and supports training and testing of those methods on different datasets using graphical user interface. The tool was evaluated with users with different level of knowledge in the machine learning field. Evaluation results confirmed that this tool is intuitive and easy to use for experts, as well as for beginners and researchers from different domains that can use GCRF for data prediction.

Brzi razvoj digitalne multimedije, dostupnost većih propusnih opsega u komunikacionim mrežama i porast procesorske snage prouzrokovali su svakodnevno korišćenje digitalnih multimedijalnih podataka na različitim uređajima i u različitim sferama života. Velika količina kako ličnih tako i poslovnih multimedijalnih podataka postaje javno dostupno i mogu biti lako ukradeni, kopirani ili modifikovani..., Following advancement and rapid development of digital multimedia, larger bandwidths available within the communication network and increased processing power led to the everyday utilization of digital multimedia on different devices and in different areas of life. A large amount of both personal and business multimedia data has consequently become publicly available, and in turn can be more easily copied or modified...

This article presents a systematic approach to defining, applying, evaluating, refining, and revising metrics for students' soft skills-their abilities like critical thinking, problem solving, leadership and responsibility, communication, and collaboration. The importance of these skills in educational and work settings is growing rapidly. While such skills are easy to notice, they are hard to measure. Metrics do exist, but vary from one case to another, and are often rather implicit and vague. Contrary to that, this article proposes the use of precisely specified, measurable, low-inference indicators (metrics) to assess soft skills. The article also introduces an open set of principles that can be used to guide the specification of concrete, evidence-based metrics for different soft skills. Two case studies are used to illustrate the approach. These case studies are part of a larger research effort that has developed an open set of metrics for different soft skills; some of them are discussed in the article extensively. Generalizing the metrics used in specific educational contexts is also discussed.

Ljudska interakcija na Vebu sadrži emocije često nepristupačne računarima...

This chapter presents the IntelLEO solution that was developed in order to provide a new paradigm, namely, Intelligent Learning Extended Organization (IntelLEO). IntelLEO leverages intelligent technologies to support learning and knowledge-building (LKB) activities of a community that emerges as a temporal integration of two or more different business and educational communities and organizational cultures, with emphasis to LKB activities of Individuals and groups across various boundaries (vertical-within the organizations involved and horizontal-within and between the organizations).. The solution is implemented as part of the IntelLEO project, an FP7 project in the area of technology enhanced learning. IntelLEO software solution supports the process of transformation between tacit and explicit knowledge on the cross-organizational level, describing the ways for improving the responsiveness of organizations participating in IntelLEO. Responsiveness in this context is defined through increased motivation to proactively learn and construct knowledge in an IntelLEO. The IntelLEO pedagogical model was derived from the initial organizational SECI knowledge management model and adapted to describe crossorganizational learning in extended organizations. The IntelLEO technological services were developed to address motivational enablers within SECI model. Thereby, the IntelLEO SECI model for promoting motivated LKB acts across organizations was supported with the technological means that enabled to apply the model in extended organizations.