• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.3
• /
• pp.411-425
• /
• 2023

Today, AI (Artificial Intelligence) technology is widely used in various fields, performing classification and regression tasks according to the purpose of use, and research is also actively progressing. Especially in the field of security, unexpected threats need to be detected, and unsupervised learning-based anomaly detection techniques that can detect threats without adding known threat information to the model training process are promising methods. However, most of the preceding studies that provide interpretability for AI judgments are designed for supervised learning, so it is difficult to apply them to unsupervised learning models with fundamentally different learning methods. In addition, previously researched vision-centered AI mechanism interpretation studies are not suitable for application to the security field that is not expressed in images. Therefore, In this paper, we use a technique that provides interpretability for detected anomalies by searching for and comparing optimization references, which are the source of intrusion attacks. In this paper, based on reference, we propose additional logic to search for data closest to real data. Based on real data, it aims to provide a more intuitive interpretation of anomalies and to promote effective use of an anomaly detection model in the security field.

• 한국정보교육학회:학술대회논문집
• /
• 2021.08a
• /
• pp.213-221
• /
• 2021

With the advent of the 4th industrial revolution, the need for artificial intelligence education has increased. The online learning environment caused by COVID-19 made it possible to use variety of artificial intelligence platforms. In this study, an aritificial intelligence class plan was developed and proposed to achieve the goal of artificial intelligence education using an AI platform. The AI platform used is AI for Oceans, With the theme of creating a program for the environment, designed a 6-hour project class using Novel Engineering-based on STEAM model. Students experience AI for Oceans enough time and learn supervised learning by experience. Based on understanding of supervised learning, students design their own programs for the environment using Entry's AI blocks. In this study, for AI convergence education, this lesson was developed and presented with the goal of acquiring the creative problem solving ability and integrated thinking ability by using the principles of artificial intelligence to solve problems.

• Journal of The Korean Association For Science Education
• /
• v.42 no.1
• /
• pp.97-109
• /
• 2022

In this study, we investigated pre-service teachers' reflection of questions during science instruction in teaching practicum from the perspective of productive reflection. The productive reflection used in this analysis has four aspects of learners and learning, subject matter knowledge, instruction, and assessment. Five pre-service teachers participated in this study. They reflected on their questions with one class video by using think-aloud method. Semi-constructed interviews were also conducted. The analyses of the results reveal that the aspect of 'instruction' and the 'learners and learning' were frequently included in their reflections. 'subject matter knowledge' was often included in their reflection while 'assessment' was hardly included. The integrations of the two aspects appear most often, those of three aspects appear only sometimes. However, four aspects appear very rarely. The integrations of 'learners and learning' and 'instruction' with the other aspects were most frequent, and the integrations of 'content knowledge' with the other aspects were often. However, the integration of 'assessment' was very few. There were more productive reflections from pre-service teachers who reflect on several questions in context than who reflected on questions one by one. In addition, they exhibited some difficulties in suggesting alternatives to improve their questions. They tried to modify the teaching method related to the questions rather than revise the form of questions. Based on the results, practical implications to improve expertise related to the questions of pre-service teachers were discussed.

• Education of Primary School Mathematics
• /
• v.25 no.4
• /
• pp.459-475
• /
• 2022

The equal sign and equivalence are the most basic and core concepts in elementary mathematics, but there has been lack of research on how to teach these concepts with textbooks. Given this, this study analyzed elementary mathematics textbooks in terms of three instructional elements (i.e., emphasizing the meaning of the equal sign as a relational symbol, dealing with an equation as an object for reasoning, and using an equation with a missing value). In particular, this study analyzed 10 different mathematics textbook series that are newly used in 2022 and examined the overall trends and characteristics for teaching the equal sign and equivalence. The results of this study showed that the activities emphasizing the meaning of the equal sign as a relational symbol were most noticeable but the activities dealing with an equation as an object for reasoning or using an equation with a missing value were relatively rare. Based on the results of the analysis, this study provides textbook writers with implications on what to further consider in covering the equal sign and equivalence.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.9
• /
• pp.407-418
• /
• 2023

The Multi-Object Goal Visual Navigation(MultiOn) is a visual navigation task in which an agent must visit to multiple object goals in an unknown indoor environment in a given order. Existing models for the MultiOn task suffer from the limitation that they cannot utilize an integrated view of multimodal context because use only a unimodal context map. To overcome this limitation, in this paper, we propose a novel deep neural network-based agent model for MultiOn task. The proposed model, MCFMO, uses a multimodal context map, containing visual appearance features, semantic features of environmental objects, and goal object features. Moreover, the proposed model effectively fuses these three heterogeneous features into a global multimodal context map by using a point-wise convolutional neural network module. Lastly, the proposed model adopts an auxiliary task learning module to predict the observation status, goal direction and the goal distance, which can guide to learn the navigational policy efficiently. Conducting various quantitative and qualitative experiments using the Habitat-Matterport3D simulation environment and scene dataset, we demonstrate the superiority of the proposed model.

• Communications of Mathematical Education
• /
• v.38 no.2
• /
• pp.145-165
• /
• 2024

The purpose of this study is to analyze the types of errors made by Korean language learners in the use of dual numerals and provides basic data for developing an effective teaching numeration using dual numerals. To this end, a case study was conducted to analyze the types of errors that appear in numeration using dual numerals targeting Korean language learners with diverse linguistic and cultural backgrounds and different academic achievements in Korean and mathematics. Error types that categorized errors made by Korean language learners were used as an analysis framework. The conclusions obtained from the research results are as follows. First, it is necessary to provide students with opportunities to use them frequently so that they can become familiar with the use of native language numerals, which often causes errors. Second, when teaching Korean language learners with low-level Korean language academic achievement how to use Chinese numerals, it is necessary to pay attention to the multiplicative numeral system of Chinese numerals. Third, it is necessary to teach children to accurately read foreign word classifiers used with Chinese numerals accurately in Korean and distinguish between the classifiers 'o'clock' and 'hours'. There is a need to provide guidance so that native language/Chinese numerals can be used appropriately in succession along with Chinese classifiers. The results of this study may contribute to the development of an effective teaching numeration using dual numerals for Korean language learners with diverse linguistic and cultural backgrounds.

• Smart Media Journal
• /
• v.12 no.8
• /
• pp.45-59
• /
• 2023

Users who want to experience the metaverse for tourism are exposed to strategic planning in space for the purpose of cultural content. In addition, users learn integrated cultural content in the process of proceeding according to the virtual environment. and Along with the meaning of time and space, users will experience space-telling. It is important to induce interest from the beginning of the experience to continue the experience. However, obstacles arise in this process. This is because developers should promote connections with new information to users who do not have sufficient prior knowledge and only have keywords of interest. Therefore, efficient design methods to enhance interest should be studied in advance. But so far, there has been no research on how to systematically design prior organizers to induce interest in virtual space. This study is an interest-inducing design method that occurs in the process of developing the meaning of virtual space and storytelling of cultural content, and can be seen as a basic study using conceptual guidance-based prior organizer education and learning techniques. First, virtual space elements and human behavior theories were considered. Subsequently, five representative examples of previous organizers currently used were explored, and redesigned and proposed based on a conceptual map for information access and delivery purposes. Through this research process, it was possible to confirm that spatial attributes and cognitive interest elements were effectively transmitted to meaningful learning leading to storytelling learning and elements of service design design method through conceptual guidance.

• Journal of Educational Research in Mathematics
• /
• v.14 no.1
• /
• pp.39-69
• /
• 2004

This paper investigates the teaching and learning of Linear function relating functional contexts and suggests the improved methods of representation-shift through this analysis. The methods emphasize the link between students' preacquired knowledge of mathematical representations and the way of using those. This methods are explanatory teaching, teaching and teaming based on modelling perspectives or tasks (interpretation, prediction, translation and scaling). We categorize the 8th grade middle school students' errors on the linear function relating real contexts and make a comparative study of the error-remedial effects and the teaching and teaming methods. We present the results of a study in which representation-shift methods based on modelling perspectives and tasks are more effective in terms of flexible connection of representations and error remediation. Also, We describe how students used modelling perspective-taking to explain and justify their conceptual models, to assess the quality of their models and to make connection to other mathematical representation during the problem solving focusing on the students' self-diagnosis.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.250-253
• /
• 2017

본 논문에서는 동시적 위치 추정 및 지도 작성 (simultaneous localization and mapping)에서 루프 폐쇄 검출을 딥러닝 방법의 일종인 variational autoencoder 를 이용하여 수행하는 방법에 대해 살펴본다. Autoencoder 는 비감독 학습 방법의 일종으로 입력 영상이 신경망을 통과하여 얻은 출력 영상과 동일하도록 신경망을 학습시키는 모델이다. 이 때 autoencoder 중간의 병목 지역을 통과함에도 불구하고 입력과 동일한 영상을 계산해야 하는 제약조건이 있기 때문에 이는 차원 축소나 데이터 추상화의 목적으로 많이 사용된다. 여기서 한 단계 더 발전된 variational autoencoder 는 기존의 autoencoder 가 가진 단점인 입력 변수의 분포와 잠재 변수의 분포 사이에 상관관계가 없다는 단점을 해결하기 위해 Kullback-Leibler divergence 를 활용한 손실 함수를 정의하여 사용했다. 실험결과에서는 루프 폐쇄 검출에서 많이 사용되는 City-Centre 와 New College 데이터 집합을 사용하여 평가하였으며 루프 폐쇄 검출의 결과는 정밀도와 재현율을 계산하여 나타냈다.

• Smart Media Journal
• /
• v.5 no.2
• /
• pp.8-14
• /
• 2016

In this paper, we propose a model for coaching and formal operating procedures performed capstone design advisor during the design done to allow effective guidance and management. Coaching model is proposed through coaching and capstone design proposed by coaching on the procedure for the implementation and management, Mokpo National University Department of Computer Engineering student in the program outcomes, performance assessment of the effective operation of the capstone design based social network service the demonstrated effectiveness of the coaching model.