• Journal of Convergence for Information Technology
• /
• v.9 no.5
• /
• pp.130-136
• /
• 2019

Recently, the importance of creativity and problem-solving skills are being emphasized in engineering education. In particular, research is actively being conducted on learning models considering practicality or applicability in practice and education among many creative problem-solving methods. The objective of the present study is to develop a teaching and learning model and verify its effects in order to promote creative thinking and problem-solving skills using the TRIZ Contradiction Analysis. This study led the participants to obtain basic knowledge of creative engineering design through the creative engineering design course for freshmen at an engineering college, and come up with ideas and solutions using the TRIZ Contradiction Analysis. A survey was conducted and analyzed to verify the effectiveness of education using the proposed teaching and learning model, and as a result, the effectiveness of education has been proven by an average of 89 positive responses. Follow-up research is needed on improved application models so that the proposed learning model can be applied to various subjects.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.159-172
• /
• 2023

In general, social problem-solving research aims to create important social value by offering meaningful answers to various social pending issues using scientific technologies. Not surprisingly, however, although numerous and extensive research attempts have been made to alleviate the social problems and issues in nation-wide, we still have many important social challenges and works to be done. In order to facilitate the entire process of the social problem-solving research and maximize its efficacy, it is vital to clearly identify and grasp the important and pressing problems to be focused upon. It is understandable for the problem discovery step to be drastically improved if current social issues can be automatically identified from existing R&D resources such as technical reports and articles. This paper introduces a comprehensive dataset which is essential to build a machine learning model for automatically detecting the social problems and solutions in various national research reports. Initially, we collected a total of 700 research reports regarding social problems and issues. Through intensive annotation process, we built totally 24,022 sentences each of which possesses its own category or label closely related to social problem-solving such as problems, purposes, solutions, effects and so on. Furthermore, we implemented four sentence classification models based on various neural language models and conducted a series of performance experiments using our dataset. As a result of the experiment, the model fine-tuned to the KLUE-BERT pre-trained language model showed the best performance with an accuracy of 75.853% and an F1 score of 63.503%.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.272-275
• /
• 2019

단어 중의성 해소 방법은 지식 정보를 활용하여 문제를 해결하는 지식 기반 방법과 각종 기계학습 모델을 이용하여 문제를 해결하는 지도학습 방법이 있다. 지도학습 방법은 높은 성능을 보이지만 대량의 정제된 학습 데이터가 필요하다. 반대로 지식 기반 방법은 대량의 정제된 학습데이터는 필요없지만 높은 성능을 기대할수 없다. 최근에는 이러한 문제를 보완하기 위해 지식내에 있는 정보와 정제된 학습데이터를 기계학습 모델에 학습하여 단어 중의성 해소 방법을 해결하고 있다. 가장 많이 활용하고 있는 지식 정보는 상위어(Hypernym)와 하위어(Hyponym), 동의어(Synonym)가 가지는 의미설명(Gloss)정보이다. 이 정보의 표상을 기존의 문장의 표상과 같이 활용하여 중의성 단어가 가지는 의미를 파악한다. 하지만 정확한 문장의 표상을 얻기 위해서는 단어의 표상을 잘 만들어줘야 하는데 기존의 방법론들은 모두 문장내의 문맥정보만을 파악하여 표현하였기 때문에 정확한 의미를 반영하는데 한계가 있었다. 본 논문에서는 의미정보와 문맥정보를 담은 단어의 표상정보를 만들기 위해 구문정보, 의미관계 그래프정보를 GCN(Graph Convolutional Network)를 활용하여 임베딩을 표현하였고, 기존의 모델에 반영하여 문맥정보만을 활용한 단어 표상보다 높은 성능을 보였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.51-53
• /
• 2000

기존의 진화 연산의 한계를 극복하기 위해서 탐색점 분포 학습 알고리즘(Estimation of Distribution Algorithm)이 부각되고 있다. 탐색점 분포 학습 알고리즘은 데이터의 분포를 파악하고, 파악된 분포를 이용해서 새로운 학습 데이터를 생성하는 일련의 과정을 통하여 최적화 문제를 해결하는 방법이다. 그런데, 기존의 탐색점 분포 학습 알고리즘들은 대부분 이진 벡터값을 가지는 최적화 문제들만을 대상으로 하고 있다. 본 논문에서는 비감독 확률 신경망 모델인 헬름홀츠 머신을 이용해서 데이터의 분포를 학습하여 연속 함수 최적화 문제를 해결하는 방법을 개발하였다. 테스트 함수들에 대해서 실수 표현형을 사용한 유전자 알고리즘과 결과를 비교하여 제안하는 방법의 우수성을 검증하였다.

• Journal of Digital Convergence
• /
• v.19 no.7
• /
• pp.51-56
• /
• 2021

This paper studies VPBL(Various PBL) models, applies them to classes, and analyzes results so that students of various majors can design and implement problems according to the characteristics of their majors in order to improve problem solving in education. VPBL performs the process of designing and implementing problems that reflect the characteristics of the major by applying constraints to the professor's programming language. The professor performs mini_class in the process of solving the designed problem and then shares it throughout. VPBL model apply results, The traditional teaching method was 3.51 points and the application of the VPBL model was 4.52 points, and "interaction, understanding of learning contents, and acquiring knowledge related to curriculum" were improved. In addition, VPBL has the advantage of expanding the learning range in the solving process as it is based on various problem solving, which has the effect of expanding the learning range compared to existing class models. Research on the expanded application of various SW liberal education in the future is required.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.881-882
• /
• 2020

최근 딥 러닝 (deep learning) 기술의 큰 발전으로 기존 기계 학습 분야의 기술들이 성공적으로 해결하지 못하던 많은 문제들을 해결할 수 있게 되었다. 이러한 딥 러닝의 학습 과정은 매우 많은 연산을 요구하기에 다수의 노드들로 모델을 학습하는 분산 학습 (distributed training) 기술이 연구되었다. 대표적인 분산 학습 기법으로 파라미터 서버 기반의 분산 학습 기법들이 있으며, 이 기법들은 파라미터 서버 노드가 학습의 병목이 될 수 있다는 한계를 갖는다. 본 논문에서는 이러한 파라미터 서버 병목 문제를 해결하는 파라미터 샤딩 기법에 대해 소개하고, 각 기법 별 학습 성능을 비교하고 그 결과를 분석하였다.

• Journal of The Korean Association of Information Education
• /
• v.17 no.3
• /
• pp.233-242
• /
• 2013

"Problem-solving methods and procedures" sections in the 2009 revised informatics curriculum emphasized active use of algorithmic thinking to solve problems. And it is proposed to solve the various problems of real life using programming language for the implementation of the algorithm. Recently, various Educational Programming Language has been developed for elementary programming activity and many researches showed that students' cognitive burden was reduced in learning programming language with Educational Programming Languages. However implementation of the algorithm is difficult for novice programmer. For the reason, effective way is required for elementary students to connect design of the algorithm and implementation of the algorithm. Therefore, in this study propose the algorithm design worksheets that it is possible to create an algorithm to describe the content needed to implementation in programming education. And this study proved the effect of the algorithm design learning tools through experiment.

• The KIPS Transactions:PartA
• /
• v.19A no.3
• /
• pp.147-154
• /
• 2012

The higher thinking skills, such as creativity and problem-solving about a given problem, are difficult to assess and diagnose. For an accurate diagnosis of these higher thinking abilities, we need to fully observe learner's problem-solving process or learner's individual reports. However, in an online learning or virtual class environments, evaluation of learner's problem-solving process becomes more difficult to diagnose. The best way to solve this problem is through reporting by tracking learner's actions when he tries to solve a problem. In this study, we developed a module which can evaluate and diagnose student's problem-solving ability by tracking actions in MS-Office suite, which is used by students to solve a given problem. This module performs based on the learner's job history through user tracking. To evaluate the effectiveness of this diagnostic module, we conducted satisfaction survey from students who were preparing the actual MOS exams. As a result, eighty-one (81) of the participants were positive on the effectiveness of the learning system with the use of this module.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.49-56
• /
• 2023

The purpose of this study is to develop convergence major-based learning (CMBL), which selects performance tasks related to local problems at hand and solves them based on convergence major performance, and builds a suitable teaching and learning model. We developed a CMBL class with a team project-type class that finds and solves practical problems in the region to cultivate overall problem-solving capabilities for convergence major competencies. Additionally, for this class, the instructor played a role as a bridgehead to explore and connect the community's sites, and students visited connected institutions in person to identify problems they need based on understanding and empathy for the subjects through field observation and qualitative interviews, and developed a CMBL class teaching and learning model necessary to directly solve them by using their major capabilities to the fullest. Therefore, we intend to present the future-oriented direction of university convergence education required by the community by forming a group of students with various majors to cultivate the ability to solve realistic problems in the community.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.2
• /
• pp.65-72
• /
• 2021

Recently, attacks on the network environment have been rapidly escalating and intelligent. Thus, the signature-based network intrusion detection system is becoming clear about its limitations. To solve these problems, research on machine learning-based intrusion detection systems is being conducted in many ways, but two problems are encountered to use machine learning for intrusion detection. The first is to find important features associated with learning for real-time detection, and the second is the imbalance of data used in learning. This problem is fatal because the performance of machine learning algorithms is data-dependent. In this paper, we propose the HSF-DNN, a network intrusion detection model based on a deep neural network to solve the problems presented above. The proposed HFS-DNN was learned through the NSL-KDD data set and performs performance comparisons with existing classification models. Experiments have confirmed that the proposed Hybrid Feature Selection algorithm does not degrade performance, and in an experiment between learning models that solved the imbalance problem, the model proposed in this paper showed the best performance.