• 한국컴퓨터정보학회논문지
• /
• 제22권11호
• /
• pp.125-133
• /
• 2017

Recently, as the programming education has become essential in school, discussion of how to teach programming has been important. This study performed a meta-analysis of the effect size depending on the teaching and learning method for the programming education. 78 research data selected from 45 papers were analyzed from cognitive and affective aspects according to dependent variables. The analysis from the cognitive aspect showed that there was no statistically significant difference in the effect size depending on whether or not the teaching and learning method was specified in the research paper. Meta-analysis of the research data where the teaching and learning method was designated displayed significances in CPS, PBL and Storytelling. Unlike the cognitive aspect, the analysis from the affective aspect showed that the effect size of the research data without the specified teaching and learning method was larger than those with specified teaching and learning method with a statistical significance. Meta-analysis of the data according to the teaching and learning method displayed no statistical significance. Based upon these research results, this study suggested implications for the effective programming education.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2024년도 춘계학술발표대회
• /
• pp.711-714
• /
• 2024

In the dynamic landscape of modern machine learning, Federated Learning (FL) has emerged as a compelling paradigm designed to enhance privacy by enabling participants to collaboratively train models without sharing their private data. Specifically, Distillation-based Federated Learning, like Federated Learning with Model Distillation (FedMD), Federated Gradient Encryption and Model Sharing (FedGEMS), and Differentially Secure Federated Learning (DS-FL), has arisen as a novel approach aimed at addressing Non-IID data challenges by leveraging Federated Learning. These methods refine the standard FL framework by distilling insights from public dataset predictions, securing data transmissions through gradient encryption, and applying differential privacy to mask individual contributions. Despite these innovations, our survey identifies persistent vulnerabilities, particularly concerning the susceptibility to logit inversion attacks where malicious actors could reconstruct private data from shared public predictions. This exploration reveals that even advanced Distillation-based Federated Learning systems harbor significant privacy risks, challenging the prevailing assumptions about their security and underscoring the need for continued advancements in secure Federated Learning methodologies.

• 한국컴퓨터정보학회논문지
• /
• 제25권4호
• /
• pp.37-45
• /
• 2020

딥러닝의 개발 프로세스는 대량의 수작업이 요구되는 반복적인 작업으로 그 중 학습 데이터 전처리는 매우 큰 비용이 요구되며 학습 결과에 중요한 영향을 주는 단계이다. AI의 알고리즘 연구 초기에는 주로 데이터 과학자들에 의해 완벽하게 정리하여 제공된 공개 DB형태의 학습데이터를 주로 사용하였다. 실제 환경에서 수집된 학습 데이터는 주로 센서들의 운영 데이터이며 필연적으로 노이즈가 많이 발생할 수 있다. 따라서 노이즈를 제거하기 위한 다양한 데이터 클리닝 프레임워크와 방법들이 연구되었다. 본 논문에서는 IoT환경에서 발생 될 수 있는 센서 데이터와 같은 시계열 데이터에서 노이즈를 감지하고 제거하는 방법을 제안하였다. 이 방법은 선형회귀 방법을 사용하여 시스템이 반복적으로 노이즈를 찾아내고, 이를 대체할 수 있는 데이터를 제공하여 학습데이터를 클리닝한다. 제안된 방법의 효과를 검증하기 위해서 본 연구에서 시뮬레이션을 수행하여, 최적의 클리닝 결과를 얻을 수 있는 인자들의 결정 방법을 확인하였다.

• 디지털산업정보학회논문지
• /
• 제7권1호
• /
• pp.75-84
• /
• 2011

The e-learning contents offered in the current educational system does not appropriately reflect the needs of actual users in the planning and development phases. Considering this problem, this study sets the following four topics as its research: Stability of e-learning; Obstacles of the applications of e-learning; e-learning contents that users wants to be offered besides lectures; and methods of e-learning, and based on these goals, it aims at determining the 'needs of the users for the promotion of e-learning. As the target of the study, a survey was conducted with 200 students who have experienced taking e-learning classes in four universities located in Eastern Seoul, which have introduced an e-learning system. The data collected from the survey went through data coding and data cleaning processes and were analyzed by year, major, and department using SAS 9 statistics package program. The result of this study showed that developing and offering services of e-learning contents that are customized to students based on their majors and year can become an effective plan for promoting e-learning.

• 수산해양교육연구
• /
• 제26권1호
• /
• pp.156-165
• /
• 2014

The purpose of this study was to find research trends of smart learning. For this, we identified the research's characteristics such as the subject or keyword of research, method, data collection, and statistical analysis method. The 2,865 articles published from 1995 to 2013 were gathered from five Korean academic journals related to smart learning. Among them, research keyword, areas, research method, data collection method, and statistical analysis method were analyzed on 596 papers. The findings of this study were as follows: (a) Smart learning papers such keyword likes u-learning, m-learning, and smart-learning were emerging after 2006. Smart learning papers with ICT related topics were highly increased after 2000, but they were decreased after 2006. Smart learning papers with e-learning related keywords were steadily increased after 2000 through 2013. (b) The research field of deign had the highest portion in smart learning research, but managing had the lowest portion. (c) Development was mainly used as a research method. Both questionnaire and experiment were mainly used for collecting data methods. T-test and frequency analysis were mainly used as statistical analysis methods.

• 한국인터넷방송통신학회논문지
• /
• 제20권5호
• /
• pp.89-95
• /
• 2020

이러닝 시장이 확대됨에 따라 인공지능 기반의 학습자 맞춤형 교육에 대한 관심이 높아지고 있다. 학습자 맞춤형 교육은 학습자 분석을 위한 대량의 데이터 및 학습 콘텐츠 등의 필수 구성요소가 필요하며 이러한 데이터 수집을 위한 시간과 비용 측면의 노력이 필요하다. 본 논문에서는 소규모 학습그룹에서의 효율적으로 학습자 맞춤형 학습이 가능하도록, python 모듈들을 사용하여 비정형 학습자 데이터를 분석하였으며 이를 토대로 제시된 학습알고리즘을 통하여 학습자의 학습연속성을 유지하도록 하였다. 본 논문을 통하여 제시된 비정형 학습데이터분석을 통하여 학습관련 비정형 데이터를 정량화 하여 측정 가능하도록 하였으며 학습자 맞춤교육 제공을 위한 키워드 분석 시 90% 이상 데이터가 유의미함을 확인하였다.

• 한국컴퓨터정보학회논문지
• /
• 제26권12호
• /
• pp.69-75
• /
• 2021

최근 인공지능 분야에서 가장 많이 사용하는 딥러닝은 그 구조가 점차 크고 복잡해지고 있다. 딥러닝 모델이 커질수록 이를 학습시키기 위해서는 대용량의 데이터가 필요하지만 데이터가 여러 소유 주체별로 분산되어 있고 보안 문제로 인해 이를 통합하여 학습시키기 어려운 경우가 발생한다. 우리는 동일한 딥러닝 모형이 필요하지만 보안 문제로 인해 데이터가 여러곳에 분산되어 처리될 수 밖에 없는 상황에서 데이터를 소유하고 있는 주체별로 분산 학습을 수행한 후 이를 통합하는 방법을 연구하였다. 이를 위해 보안 상황을 V-환경과 H-환경으로 가정하여 소유 주체별로 분산학습을 수행했으며 Average, Max, AbsMax를 사용하여 분산학습된 결과를 통합하였다. mnist-fashion 데이터에 이를 적용해 본 결과 V-환경에서는 정확도 면에서 데이터를 통합시켜 학습한 결과와 큰 차이가 없음을 확인할 수 있었으며, H-환경에서는 차이는 존재하지만 의미있는 결과를 얻을 수 있었다.

• 한국간호교육학회지
• /
• 제27권4호
• /
• pp.381-391
• /
• 2021

Purpose: This study aimed to identify the influence of learning presence and self-directed learning ability on nursing students' learning satisfaction according to the online learning method. Methods: The participants of this study were 167 nursing students attending three universities in different cities. The data were collected from July 16 to July 23, 2021, via an online self-reported questionnaire. Using SPSS WIN 27.0, data were analyzed by descriptive statistics, Pearson's correlation coefficient and a multiple regression analysis. Results: The most effective online learning method experienced by nursing students was asynchronous online learning according to 58.2% of the respondents, while 30.3% of the respondents answered synchronous online learning. The main merit of asynchronous online learning was that it was possible to listen repeatedly (61.7%) to lectures, and the top advantage of synchronous online learning was that the location of the class was free (53.3%). In asynchronous online learning, the factors that significantly affected nursing students' learning satisfaction were cognitive presence (𝛽=.60, p<.001) and emotional presence (𝛽=.25, p<.001). These variables accounted for 56% of their learning satisfaction (F=54.12, p<.001). Similarly, cognitive presence (𝛽=.64, p<.001) and emotional presence (𝛽=.21, p=.001) in synchronous online learning, were the factors cited for significantly affecting learning satisfaction. The explanatory power was 62% (F=69.19, p<.001). Conclusions: In conclusion, it was found that cognitive and social presence from the learning presence factors in both asynchronous and synchronous online learning influence and enhance nursing students' learning satisfaction. Therefore, these results provide important data for future online class design in nursing education.

• 간호행정학회지
• /
• 제16권3호
• /
• pp.267-275
• /
• 2010

Purpose: This study conducted to identify the effect of learning organization construction and learning orientation on organizational effectiveness among hospital nurses. Method: Data was collected from convenient sample of 296 nurses who worked for the major hospitals in Seoul, Gyeonggi-do and Gangwoen-do. The self-reported questionnaire was used to assess the general characteristics, the level of the learning organization construction, learning orientation and organizational effectiveness. The data were analyzed using descriptive statistics, pearson's correlation coefficient and multiple regression. Result: The mean score of learning organization construction was 3.61(${\pm}.32$), learning orientation got 3.26(${\pm}.39$), and organizational effectiveness obtained 3.38(${\pm}.42$). The learning organization construction affects of organizational effectiveness by 44.18% and learning orientation by 37.43%. Conclusion: This finding indicates that learning organization construction and learning orientation affects the nurses' organizational effectiveness in hospital.

• International Journal of Advanced Culture Technology
• /
• 제10권4호
• /
• pp.412-419
• /
• 2022

The amount of digital text data is growing exponentially, and many machine learning solutions are being used to monitor and manage this data. Artificial intelligence and machine learning are used in many areas of our daily lives, but the underlying processes and concepts are not easy for most people to understand. At a time when many experts are needed to run a machine learning solution, no-code machine learning tools are a good solution. No-code machine learning tools is a platform that enables machine learning functions to be performed without engineers or developers. The latest No-Code machine learning tools run in your browser, so you don't need to install any additional software, and the simple GUI interface makes them easy to use. Using these platforms can save you a lot of money and time because there is less skill and less code to write. No-Code machine learning tools make it easy to understand artificial intelligence and machine learning. In this paper, we examine No-Code machine learning tools and compare their features.