Mustafa Murat Inceoglu

A hybrid genetic algorithm for packing in 3d with deepest bottom left with fill method

Three dimensional bin packing problems arise in industrial applications like container ship loading, pallet loading, plane cargo management and warehouse management, etc. In this paper, a hybrid genetic algorithm (GA) is used for regular 3D strip packing. The Genetic Algorithm is hybridized with the presented Deepest Bottom Left with Fill (DBLF) method. Several heuristic methods have also been used for comparison with the hybrid GA.

Diagnosis of learning styles based on active/reflective dimension of felder and silverman's learning style model in a learning management system

Learner centered education is important both in point of face to face and Web based learning. Due to this importance, diagnosis of learning styles of students in web based or web enhanced educational settings is important as well. This paper presents prediction of learning styles by means of monitoring learner interface interactions. A mathematics course executed on a learning management system (Moodle) was monitored and learning styles of the learners were analyzed in point of active/reflective dimension of Felder and Silverman Learning Styles Model. The data from learner actions were analyzed through literature based automatic student modeling. The results from Index of Learning Styles and predicted learning styles were compared. For active/reflective dimension 79.6% precision was achieved.

A Comparative Study on Neural Network Based Soccer Result Prediction

This study mainly remarks the efficiency of black-box modeling capacity of neural networks in the case of forecasting soccer match results, and opens up several debates on the nature of prediction and selection of input parameters. The selection of input parameters is a serious problem in soccer match prediction systems based on neural networks or statistical methods. Several input vector suggestions are implemented in literature which is mostly based on direct data from weekly charts. Here in this paper, two different input vector parameters have been tested via learning vector quantization networks in order to emphasize the importance of input parameter selection. The input vector parameters introduced in this study are plain and also meaningful when compared to other studies. The results of different approaches presented in this study are compared to each other, and also compared with the results of other neural network approaches and statistical methods in order to give an idea about the successful prediction performance. The paper is concluded with discussions about the nature of soccer match forecasting concept that may draw the interests of researchers willing to work in this area.

Developing Adaptive and Personalized Distributed Learning Systems with Semantic Web Supported Multi Agent Technology

The early e-learning systems were developed with the one-size-fits-all approach where the differences among the learners were disregarded and the same learning materials were supplied to each user. Nowadays, with the technological advances and the new trends in system design, the newly-developed systems take into consideration the needs, the preferences and the learning styles of the learners. As a result of this, more personalized e-learning systems have been developed. This thesis will investigate how possible technologies such as multi-agent systems and semantic web can be used to achieve more adaptive and more personalized distributed e-learning environments.

Exploring the state of the art in adaptive distributed learning environments

The use of one-size-fits-all approach is getting replaced by the adaptive, personalized perspective in recently developed learning environments. This study takes a look at the need of personalization in e-learning systems and the adaptivity and distribution features of adaptive distributed learning environments. By focusing on how personalization can be achieved in e-learning systems, the technologies used for establishing adaptive learning environments are explained and evaluated briefly. Some of these technologies are web services, multi-agent systems, semantic web and AI techniques such as case-based reasoning, neural networks and Bayesian networks used in intelligent tutoring systems. Finally, by discussing some of the adaptive distributed learning systems, an overall state of the art of the field is given with some future trends.

Establishing a K-12 Circuit Design Program

Outreach, as defined by Wikipedia, is an effort by an organization or group to connect its ideas or practices to the efforts of other organizations, groups, specific audiences, or the general public. This paper describes a computer engineering outreach project of the Department of Computer Engineering at Ege University, Izmir, Turkey, to a local elementary school. A group of 14 K-12 students was chosen by a four-stage selection method to participate in this project. This group was then taught discrete mathematics and logic design courses from the core curriculum of the Computer Engineering program. The two 11-week courses have a total of 132 contact h. The course contents are conveyed through both theoretical lessons and laboratory sessions. All of the laboratory sessions were carried out by K-12 students. Volunteer teachers from the elementary school participated in the project. The evaluations carried out during and at the end of project indicated the degree of satisfaction on the part of students and teachers. The project is still ongoing with the same methodology in its third year.

Machine learning based learner modeling for adaptive web-based learning

Especially in the first decade of this century, learner adapted interaction and learner modeling are becoming more important in the area of web-based learning systems. The complicated nature of the problem is a serious challenge with vast amount of data available about the learners. Machine learning approaches have been used effectively in both user modeling, and learner modeling implementations. Recent studies on the challenges and solutions about learner modeling are explained in this paper with the proposal of a learner modeling framework to be used in a web-based learning system. The proposed system adopts a hybrid approach combining three machine learning techniques in three stages.

Reusable learning objects (RLOs) for computer science students

The purpose of this study is to introduce an instructional technology known as the “learning object”. After a review of the literature, the designing steps of a learning object (LO) are tried to explain. As a learning object standard a few details are given about SCORM-Content Aggregation Model and common metadata elements. At the end, a case study about designing of a learning object with the subject of “Congestion Control in Computer Networks” is tried to give. The content is prepared in English and it contains 13 pictures within 22 .htm and 3 .exe files. The basic principles of the network congestion and the designed congestion control algorithms are given in the htm pages. Simulators are used to show the working way of the related algorithms step by step. The LO design is made with an open source program RELOAD Editor, according to the ADL SCORM package. The designed LO has been tested on computer engineering students and positive feedbacks are received.

An Agents and Artifacts Metamodel Based E-Learning Model to Search Learning Resources

In this paper, an e-learning model based on Agents and Artifacts (A&A) Metamodel to search learning resources from multiple sources is proposed. Multi agent system (MAS) based e-learning models with the same functionality are available in the literature. However, they are mostly developed as standalone systems that contain a single agent responsible for searching and retrieving learning resources. With the highly distributed nature of learning resources over multiple repositories, giving this responsibility to only one agent decreases scalability. The proposed model exploits the A&A Metamodel to overcome this issue. A&A Metamodel focuses on environment modeling in MAS design and models entities in the environment as artifacts, that are first class entities like agents. From the perspective of MAS based e-learning systems, learning resources are the main components in the environment that agents interact with. Thus, an efficient solution can be achieved with an e-learning model that searches learning objects by using an e-learning environment model based on A&A Metamodel. The proposed e-learning system is developed with Jason and the e-learning environment model is implemented with CArtAgO framework. Finally, current limitations and future directions of the proposed approach are discussed.

A Felder and Silverman Learning Styles Model Based Personalization Approach to Recommend Learning Objects

In this paper, a new algorithmic personalization approach based on Felder and Silverman learning styles model is presented. The proposed approach uses learning objects modeled with the IEEE LOM metadata standard, which serves as the main standard for representation of learning objects’ metadata. Personalization is provided with two steps in the proposed approach. At the first step, each learning object is evaluated by taking into account how values of IEEE LOM metadata elements match each dimension of Felder and Silverman learning styles model. The second step involves recommending appropriate learning objects to learners. Four weight values are calculated for each learning object, describing how related the learner and the learning object in question is at each dimension of Felder and Silverman learning styles model. Then, weight values for each dimension is combined by using Manhattan distance metric to provide a single weight value as a fitness function representing the general relatedness of the learner and the learning object. Results of the personalization approach can be used to recommend learning objects ordered according to their weight values to the learners. An example scenario illustrating the proposed approach is provided, as well as a discussion of current limitations and future work directions.