Search

Cilj ovoga završnog rada je demonstrirati sve sposobnosti dubinskog i strojnog učenja pomoću alata za izradu virtualnih simulacija. Unity je game engine koji se koristi za izradu trodimenzionalnih (3D) i dvodimenzionalnih (2D) igara, interaktivnih simulacija itd. Simulacija je napravljena s dodatkom ML-Agents koji omogućuje primjenu strojnog učenja i obuku agenata za obavljanje određenih zadataka. Simulacija je stvorena kao dio testiranja mogućnosti u kojem se agenta podučavaja korištenjem strojnog učenja uz podršku. Unutar određenog vremena, agent mora pronaći nasumično generiranu kombinaciju brave. Agent je uspješno savladao zadani zadatak. This final project aims to demonstrate all the potentials of deep learning and machine learning by employing tools for building virtual simulations. Three-dimensional (3D) and two-dimensional (2D) games, interactive simulations, and other experiences can all be made using the game engine Unity. The simulation was created with an add-on called ML-Agents, which enables the application of machine learning and the training of agents to carry out particular tasks. A simulation was created as part of the effort in which the agents are instructed using machine learning with support. Within a set amount of time, the agent has to figure out a randomly generated lock combination. The assigned task has been successfully completed by the agent

The field of learning design studies how to support teachers in devising suitable activities for their students to learn. The field of learning analytics explores how data about students' interactions can be used to increase the understanding of learning experiences. Despite its clear synergy, there is only limited and fragmented work exploring the active role that data analytics can play in supporting design for learning. This paper builds on previous research to propose a framework (analytics layers for learning design) that articulates three layers of data analytics-learning analytics, design analytics and community analytics - to support informed decision-making in learning design. Additionally, a set of tools and experiences are described to illustrate how the different data analytics perspectives proposed by the framework can support learning design processes. Refereed/Peer-reviewed

Current research shows that digital games can significantly enhance children's learning. The purpose of this study was to examine how design features in 12 digital math games influenced children's learning. The participants in this study were 193 children in Grades 2 through 6 (ages 8–12). During clinical interviews, children in the study completed pre-tests, interacted with digital math games, responded to questions about the digital math games, and completed post-tests. We recorded the interactions using two video perspectives that recorded children's gameplay and responses to interviewers. We employed mixed methods to analyze the data and identify salient patterns in children's experiences with the digital math games. The analysis revealed significant gains for 9 of the 12 digital games and most children were aware of the design features in the games. There were eight prominent categories of design features in the video data that supported learning and mathematics connections. Six categories focused on how the design features supported learning in the digital games. These categories included: accuracy feedback, unlimited/multiple attempts, information tutorials and hints, focused constraint, progressive levels, and game efficiency. Two categories were more specific to embodied cognition and action with the mathematics, and focused on how design features promoted mathematics connections. These categories included: linked representations and linked physical actions. The digital games in this study that did not include linked representations and opportunities for linked physical actions as design features did not produce significant gains. These results suggest the key role of mathematics-specific design features in the design of digital math games.

The authors studied primary school students’ experiences of using an electronic portfolio in their craft education over four years. A stimulated recall interview was applied to collect user experiences and qualitative content analysis to analyse the collected data. The results indicate that the electronic portfolio was experienced as a multipurpose tool to support learning. It makes the learning process visible and in that way helps focus on and improves the quality of learning. © ISLS.

Doctoral thesis titled Posibilities of support for teaching a foreign language to pupils with special educational needs at younger school age responds to the need to change the traditional approach in teaching a foreign language to pupils with specific difficulties. The aim is to map the use of a support program designed to support the curriculum in the English language to pupils with special educational needs of younger school age and on the basis of proven support program of applied within normal normal teaching hours to propose recommendations for practical language teaching. Theoretical bases of the current knowledge about specific learning difficulties young learners. Research is based on analysis of identified problem areas and analyzing pupils' acquired knowledge of the English language among pupils research sample, under which the pupils of the experimental group targeted presented support program, which includes elaborate didactic material aiming to promote the adoption of English curriculum. Analysis of the results of the output test of acquired knowledge pupils of the experimental group was focused on mapping the displacement achieved language skills, analyze the level of motivation and activities of pupils. In the end, the support program is evaluated and recommenndations are made arising from...

The basis of problem-model approach is described in this article, two basic types of training in modern pedagogy are discussed, problem-model approach is analyzed from the point of view of its effectiveness, the indicators of the problem-model approach in teaching are picked out В данной статье рассматриваются основы и истоки проблемно-модельного подхода в обучении, описываются два основных типа обучения в современной педагогике, анализируется проблемно-модельный подход с точки зрения его эффективности, выделяются признаки проблемно-модельного подхода в обучении

Recent researches in e-Learning area highlight the need to define novel and advanced support mechanism for commercial and academic organizations in order to enhance the skills of employees and students and, consequently, to increase the overall competitiveness in the new economy world. This is due to the unbelievable velocity and volatility of modern knowledge that require novel learning methods which are able to offer additional support features as efficiency, task relevance and personalization. This paper tries to deal with these features by proposing an adaptive e-Learning framework based on Computational Intelligence methodologies by supporting e-Learning systems' designers in two different aspects: (1) they represent the most suitable solution, able to support learning content and activities, personalized to specific needs and influenced by specific preferences of the learner and (2) they assist designers with computationally efficient methods to develop "in time" e-Learning environments. Our work attempts to achieve both results by exploiting an ontological representation of learning environment and a hierarchical memetic approach of optimization. In detail, our approach takes advantage of a collection of ontological models and processes for adapting an e-Learning system to the learner expectations by efficiently solving a well-defined optimization problem, through a hierarchical multi-cores memetic approach. © 2009 Springer-Verlag London Limited.

Learning is a critical support mechanism for industrial and academic organizations to enhance the skills of employees and students and, consequently, the overall competitiveness in the new economy. The remarkable velocity and volatility of modern knowledge require novel learning methods offering additional features as efficiency, task relevance and personalization. Computational Intelligence methodologies can support e-Learning system designers in two different aspects: (1) they represent the most suitable solution able to support learning content and activities, personalized to specific needs and influenced by specific preferences of the learner and (2) they assist designers with computationally efficient methods to develop "in time" e-Learning environments. This article attempts to achieve both results by exploiting an ontological representations of learning environment and memetic approach of optimization, integrated into a cooperative distributed problem solving framework. This synergy enables multi-island memetic approach managing a collection of models and processes for adapting an e-Learning system to the learner expectations and to formulate objectives in an effective and dynamic intelligent way. More precisely, our proposal exploits ontological representations of learning environment and a memetic distributed problem-solving approach to generate the best learning presentation and, at the same time, minimize the computational efforts necessary to compute optimal learning experiences. © 2011 Wiley Periodicals, Inc.