• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.453-461
• /
• 2021

Many deep learning approaches have been studied for image classification in computer vision. However, there are not enough data to generate accurate models in medical fields, and many datasets are not annotated. This study presents a new method that can use both unlabeled and labeled data. The proposed method is applied to classify cervix images into normal versus cancerous, and we demonstrate the results. First, we use a patch self-supervised learning for training the global context of the image using an unlabeled image dataset. Second, we generate a classifier model by using the transferred knowledge from self-supervised learning. We also apply attention learning to capture the local features of the image. The combined method provides better performance than state-of-the-art approaches in accuracy and sensitivity.

• International Journal of Computer Science & Network Security
• /
• v.21 no.4
• /
• pp.145-166
• /
• 2021

M-learning is one of the most important modern learning environments in developed countries, especially in the context of the COVID-19 pandemic. According to the Ministry of Education policies in Saudi Arabia, gender segregation in education reflects the country's religious values, which are a part of the national policy. Thus, it will help many in the target audience to accept online learning more easily in Saudi society. The literature review indicates the importance to use the UTAUT conceptual framework to study the level of acceptance through adding a new construct to the model which is Mobile Application Quality. The study focuses on the end user's requirements to use M-learning applications. It is conducted with a qualitative method to find out the students' and companies' opinions who working in the M-learning field to determine the requirements for the development of M-learning applications that are compatible with the aspirations of conservative societies.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.6
• /
• pp.1230-1237
• /
• 2009

In this paper, we develop a design methodology of Granular-based Radial Basis Function Neural Networks(GRBFNN) by context-based clustering. In contrast with the plethora of existing approaches, here we promote a development strategy in which a topology of the network is predominantly based upon a collection of information granules formed on a basis of available experimental data. The output space is granulated making use of the K-Means clustering while the input space is clustered with the aid of a so-called context-based fuzzy clustering. The number of information granules produced for each context is adjusted so that we satisfy a certain reconstructability criterion that helps us minimize an error between the original data and the ones resulting from their reconstruction involving prototypes of the clusters and the corresponding membership values. In contrast to "standard" Radial Basis Function neural networks, the output neuron of the network exhibits a certain functional nature as its connections are realized as local linear whose location is determined by the values of the context and the prototypes in the input space. The other parameters of these local functions are subject to further parametric optimization. Numeric examples involve some low dimensional synthetic data and selected data coming from the Machine Learning repository.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.4
• /
• pp.423-429
• /
• 2006

The systems in the ubiquitous computing environment need to provide users with context-aware services, intelligently interacting with the surrounding environment. Therefore, the systems in the ubiquitous computing environment require context-awareness ability in order to gather and analyze context information in various situations and environments. However, existing context-aware systems lack the ability to systematically generate and handle various types of context information, and only a few systems have ability learning from environment. In this paper, a general context model is defined to describe various contexts and a context-awareness framework is implemented based in the model, which makes it straightforward to handle and generate various types of context from diverse sensor. The framework is designed to allow a system to sensed, combined, inferred, and learned context information, in order to provide users with services in dynamic environments. We have implemented the proposed framework and applied it to a u-Health management system.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.11a
• /
• pp.205-210
• /
• 2005

유비쿼터스 컴퓨링과 네트워킹 환경이 준비됨에 따라 교육 분야에서도 새로운 환경에 적합한 교육학습 시스템에 대한 준비가 필요하다. 특히 유비쿼터스 컴퓨팅 환경에서는 단순히 새로운 기술을 교육학습 분야에 적용하는 것이 아니라 사고방식과 대상을 바꾸는 패러다임의 전환이 필요하다. 분야에서는 유비쿼터스 환경을 단계적으로 적용하여야 한다. 기존의 e-learning에서는 지능시스템이 교육학습 분야에 적용될 수 있는 부분이 한정되어 있었다. 그러나 유비쿼터스 맞춤형 학습 시스템을 구축할 수 있는 기본 환경이 제공하기 위하여 유비퀴터스 환경의 하부 단위에서 증강현실(augmented reality) 기술, 지능형 학습 기술들을 도출하고 적용 방법을 제안한다.

• International Journal of Computer Science & Network Security
• /
• v.22 no.12
• /
• pp.116-124
• /
• 2022

Information and communication technologies used in the social sphere are born due to the development of computer technologies. The main task of the distance learning process in higher education institutions is not to provide information, but to teach how to obtain and use it. The purpose of the article: to identify innovative approaches in the training of specialists in higher education institutions in the context of distance learning. Various innovative approaches to organizing the work of students of higher educational institutions in the context of distance learning are considered. Based on the conducted research, it is concluded that each of the approaches described by us outlines the study of the phenomenon of professional training of a specialist in the condition of distance learning. All the described approaches significantly contribute to the improvement of professional training of specialists, encourage students to self-improvement, professional development and enrich their professional competence in modern conditions. The emergence and spread of innovative technologies means not only a change in the activity itself and its inherent means and mechanisms of its implementation, but also a significant restructuring of goals, value orientations, specific knowledge, skills and abilities. Therefore, the current stage of the development of civilization, scientific and technological progress requires the emergence of such specialists who would have broad humanitarian thinking, would have good psychological training, would be able to build professional activities according to laws that take into account the relationship between economic productivity and creativity, as well as the desire of the individual for constant renewal, self-realization. Only such qualities will help you master the specifics of innovative technologies well. We see the prospects in the study of innovative approaches to training specialists in higher education institutions in the condition of distance learning in foreign countries.

• Journal of Broadcast Engineering
• /
• v.18 no.3
• /
• pp.401-408
• /
• 2013

In this paper, we propose a method that automatically detects organs having similar appearances in medical images by learning both context and appearance features. Since only the appearance feature is used to learn the classifier in most existing detection methods, detection errors occur when the medical images include multiple organs having similar appearances. In the proposed method, based on the probabilities acquired by the appearance-based classifier, new classifier containing the context feature is created by iteratively learning the characteristics of probability distribution around the interest voxel. Furthermore, both the efficiency and the accuracy are improved through 'region based voting scheme' in test stage. To evaluate the performance of the proposed method, we detect femur and tibia which have similar appearance from SKI10 knee joint dataset. The proposed method outperformed the detection method only using appearance feature in aspect of overall detection performance.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.20 no.4
• /
• pp.85-100
• /
• 2009

This study suggested the expansion strategies of school library catalog considering educational context which should be used teaching and learning process. To achieve the purpose of research, this study derived educational context categories by comparing and analyzing teaching and learning related factors, information resource related factors. Also, this study analysed case system considering educational context. Based on the results, this study designed the catalog data elements as an element to be added to an existing school libraries system(DLS). The derived data element is end user(teacher, students), instructional situations (teaching method, instructional object, curriculum, evaluation type), resource type(feature, discipline, format), reading situation(contextual reading, literature topic), related materials(teacher representation, student representation).

• International Journal of Contents
• /
• v.5 no.1
• /
• pp.1-8
• /
• 2009

The new paradigm called the web 2.0 recently appeared in the web environment. We pay attention to the positive effects which may be brought about by application of the web 2.0 to e-learning; we think that it can improve problem solving skills of learners and reinforce their creativity. But until now, e-learning model, which understood the web 2.0 concept completely, has been never developed. In this context, we propose the web 2.0-based e-learning platform which induces all the courses for education such as the selection of topic, preparation of lecture schedule and contents, teaching and learning, to be decided by participants. We believe that this platform can replace or supplement the e-learning of web 1.0 age, and realize the positive effects.

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.169-180
• /
• 2021

Since machine learning was invented, there have been many different machine learning-based algorithms, from shallow learning to deep learning models, that provide solutions to the classification tasks. But then it poses a problem in choosing a suitable classification algorithm that can improve the classification/detection efficiency for a certain network context. With that comes whether an algorithm provides good performance, why it works in some problems and not in others. In this paper, we present a data-centric analysis to provide a way for selecting a suitable classification algorithm. This data-centric approach is a new viewpoint in exploring relationships between classification performance and facts and figures of data sets.