• 한국경제지리학회지
• /
• 제25권1호
• /
• pp.171-181
• /
• 2022

부동산 시장 분석에 있어 기본이 되는 정량적 데이터는 부동산 가격 지수이다. OECD와 같은 국제기구에서는 국가별 부동산 가격 지수를 공표하고, 한국부동산원에서는 광역시 단위와 시군구 단위의 지수를 산출한다. 그런데 공간단위를 시군구보다 정교한 동단위, 아파트 단지 단위로 설정하는 경우, 여러 문제점을 맞이하게 된다. 대표적인 문제는 결측치이다. 공간적 범위를 좁힐수록 단위 기간에 따라 거래가 적거나 아예 존재하지 않는 경우가 존재하기에 이 경우에는 지수의 산출이 불가능한 결측치가 발생할 수 있다. 본 연구에서는 지도학습 기반의 머신러닝 기법을 활용하여 특정 범위와 기간에 거래가 존재하지 않아 발생할 수 있는 결측치를 보완하는 기법을 제안한다. 본 모형을 통해 부동산 매매 지수의 실제값이 존재하는 것들의 예측을 통해 그 정확도를 검증하고 결측치가 발생한 것들의 예측도 해 볼 수 있었다.

• 문화기술의 융합
• /
• 제8권4호
• /
• pp.347-353
• /
• 2022

외식업은 소비자의 수요가 많고 진입장벽이 낮아 창업이 활발하게 일어난다. 하지만 외식업은 폐업률이 높고, 프랜차이즈의 경우 동일 브랜드 내에서도 매출 편차가 크게 나타난다. 따라서 외식업 프랜차이즈의 폐업을 방지하기 위한 연구가 필요하다. 이를 위해, 본 연구에서는 프랜차이즈 가맹점 매출에 영향을 미치는 요인들을 살펴보고, 도출된 요인들에 머신러닝 기법을 활용하여 프랜차이즈의 성패를 예측하고자 한다. 강남구 프랜차이즈 매장의 PoS(Point of Sale) 데이터와 공공데이터를 활용하여 가맹점 매출에 영향을 미치는 여러 요인들을 추출하고, VIF(Variance Inflation Factor)를 활용하여 다중공산성을 제거하여 타당성 있는 변수 선택을 진행한 뒤, 머신러닝 기법 중 분류모델을 활용하여 프랜차이즈 매장의 성패 예측을 진행한다. 이를 통해 최고 정확도 0.92를 가진 프랜차이즈 성패 예측 모델을 제안한다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제27권3호
• /
• pp.1-11
• /
• 2021

저속 동조는 촬영자가 장노출과 카메라 플래시를 동시에 이용해서 전경과 배경을 밝게 하는 촬영 기법이다. 단노출 플래시 촬영과 플래시 없는 장노출 촬영과는 달리 저속 동조는 어두운 환경에서의 밝은 전경과 배경을 보장한다. 하지만 스마트폰으로 저속 동조 촬영은 어려운데, 이는 스마트폰 카메라의 플래시는 약한 지속 광이고 노출 시간이 길어지면 플래시를 켜지 못하기 때문이다. 본 연구에서는 단노출 플래시 영상에서 저속 동조 영상을 만드는 딥러닝 방법을 제안한다. 본 연구에서는 공간상에서 가변적인 영상 밝기 개선을 위해 가중치 맵을 적용한 네트워크를 제안한다. 본 연구에서는 지도 학습을 위한 스마트폰 단노출 플래시 영상과 저속 동조 영상 데이터 세트도 제안한다. RAW 영상의 선형성을 이용해 단노출 플래시 영상과 플래시 없는 장노출 영상으로부터 저속 동조 영상을 생성해서 데이터 세트를 구축한다. 실험을 통해 본 연구의 방법이 저속 동조 영상을 효과적으로 생성하는 것을 볼 수 있다.

• 한국의류학회지
• /
• 제46권1호
• /
• pp.165-181
• /
• 2022

The fashion industry is creating innovative business models using artificial intelligence. To efficiently utilize artificial intelligence (AI), fashion data must be classified. Until now, such data have been classified focusing only on the objective properties of fashion products. Their subjective attributes, such as fashion brand sensibilities, are holistic and heuristic intuitions created by a combination of design elements. This study aims to improve the performance of collaborative filtering in the fashion industry by extracting fashion brand sensibility using computer vision technology. The image data set of fashion brand sensibility consists of high-end fashion brand photos that share sensibilities and communicate well in fashion. About 26,000 fashion photos of 11 high-end fashion brand sensibility labels have been collected from the 16FW to 21SS runway and 50 years of US Vogue magazines beginning from 1971. We use EfficientNet-B1 to establish the main architecture and fine-tune the network with ImageNet-ILSVRC. After training fashion brand sensibilities through deep learning, the proposed model achieved an F-1 score of 74% on accuracy tests. Furthermore, as a result of comparing AI machine and human experts, the proposed model is expected to be expanded to mass fashion brands.

• 한국군사과학기술학회지
• /
• 제25권6호
• /
• pp.617-627
• /
• 2022

Military training is an essential item for the fundamental problem of war. However, there has always been a problem that many resources are consumed, causing spatial and environmental pollution. The concepts of defense modeling and simulation and CGF(Computer Generated Force) using computer technology began to appear to improve this problem. The Naval Operations, Resources Analysis Model(NORAM) developed by the Republic of Korea Navy is also a DEVS(Discrete Event Simulation)-based naval virtual force analysis model. The current NORAM is a battle experiment conducted by an operator, and parameter values such as maneuver and armament operation for individual objects for each situation are evaluated. In spite of our research conducted evolutionary, supervised, reinforcement learning, in this paper, we introduce our design of a scenario creation model based on evolutionary learning using genetic algorithms. For verification, the NORAM is loaded with our model to analyze wartime engagements. Human-level tactical scenario creation capability is secured by automatically generating enemy tactical scenarios for human-designed Blue Army tactical scenarios.

• 정보보호학회논문지
• /
• 제33권3호
• /
• pp.411-425
• /
• 2023

오늘날 AI(Artificial Intelligence) 기술은 다양한 분야에서 활용 목적에 맞게 분류, 회기 작업을 수행하며 광범위하게 활용되고 있으며, 연구 또한 활발하게 진행 중인 분야이다. 특히 보안 분야에서는 예기치 않는 위협을 탐지해야 하며, 모델 훈련과정에 알려진 위협 정보를 추가하지 않아도 위협을 탐지할 수 있는 비 지도학습 기반의 이상 탐지 기법이 유망한 방법이다. 하지만 AI 판단에 대한 해석 가능성을 제공하는 선행 연구 대부분은 지도학습을 대상으로 설계되었기에 학습 방법이 근본적으로 다른 비 지도학습 모델에 적용하기는 어려우며, Vision 중심의 AI 매커니즘 해석연구들은 이미지로 표현되지 않는 보안 분야에 적용하기에 적합하지 않다. 따라서 본 논문에서는 침해공격의 원본인 최적화 Reference를 탐색하고 이와 비교함으로써 탐지된 이상에 대한 해석 가능성을 제공하는 기법을 활용한다. 본 논문에서는 산출된 Reference를 기반으로 실존 데이터에서 가장 가까운 데이터를 탐색하는 로직을 추가 제안함으로써 실존 데이터를 기반으로 이상 징후에 대한 더욱 직관적인 해석을 제공하고 보안 분야에서의 효과적인 이상 탐지모델 활용을 도모하고자 한다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 추계학술발표대회
• /
• pp.942-943
• /
• 2021

The present work proposes a classification system for the HEp-2 cell images using an unsupervised deep feature learning method. Unlike most of the state-of-the-art methods in the literature that utilize deep learning in a strictly supervised way, we propose here the use of the deep convolutional autoencoder (DCAE) as the principal feature extractor for classifying the different types of the HEp-2 cell images. The network takes the original cell images as the inputs and learns to reconstruct them in order to capture the features related to the global shape of the cells. A final feature vector is constructed by using the latent representations extracted from the DCAE, giving a highly discriminative feature representation. The created features will be fed to a nonlinear classifier whose output will represent the final type of the cell image. We have tested the discriminability of the proposed features on one of the most popular HEp-2 cell classification datasets, the SNPHEp-2 dataset and the results show that the proposed features manage to capture the distinctive characteristics of the different cell types while performing at least as well as the actual deep learning based state-of-the-art methods.

• International Journal of Computer Science & Network Security
• /
• 제23권7호
• /
• pp.155-164
• /
• 2023

Machine learning with its high precision algorithms, Precision agriculture (PA) is a new emerging concept nowadays. Many researchers have worked on the quality and quantity of PA by using sensors, networking, machine learning (ML) techniques, and big data. However, there has been no attempt to work on trends of artificial intelligence (AI) techniques, dataset and crop type on precision agriculture using internet of things (IoT). This research aims to systematically analyze the domains of AI techniques and datasets that have been used in IoT based prediction in the area of PA. A systematic literature review is performed on AI based techniques and datasets for crop management, weather, irrigation, plant, soil and pest prediction. We took the papers on precision agriculture published in the last six years (2013-2019). We considered 42 primary studies related to the research objectives. After critical analysis of the studies, we found that crop management; soil and temperature areas of PA have been commonly used with the help of IoT devices and AI techniques. Moreover, different artificial intelligence techniques like ANN, CNN, SVM, Decision Tree, RF, etc. have been utilized in different fields of Precision agriculture. Image processing with supervised and unsupervised learning practice for prediction and monitoring the PA are also used. In addition, most of the studies are forfaiting sensory dataset to measure different properties of soil, weather, irrigation and crop. To this end, at the end, we provide future directions for researchers and guidelines for practitioners based on the findings of this review.

• International Journal of Computer Science & Network Security
• /
• 제24권2호
• /
• pp.79-88
• /
• 2024

The use of the social media has become part of our daily life activities. The social web channels provide the content generation facility to its users who can share their views, opinions and experiences towards certain topics. The researchers are using the social media content for various research areas. Sentiment analysis, one of the most active research areas in last decade, is the process to extract reviews, opinions and sentiments of people. Sentiment analysis is applied in diverse sub-areas such as subjectivity analysis, polarity detection, and emotion detection. Stance classification has emerged as a new and interesting research area as it aims to determine whether the content writer is in favor, against or neutral towards the target topic or issue. Stance classification is significant as it has many research applications like rumor stance classifications, stance classification towards public forums, claim stance classification, neural attention stance classification, online debate stance classification, dialogic properties stance classification etc. This research study explores different feature sets such as lexical, sentiment-specific, dialog-based which have been extracted using the standard datasets in the relevant area. Supervised learning approaches of generative algorithms such as Naïve Bayes and discriminative machine learning algorithms such as Support Vector Machine, Naïve Bayes, Decision Tree and k-Nearest Neighbor have been applied and then ensemble-based algorithms like Random Forest and AdaBoost have been applied. The empirical based results have been evaluated using the standard performance measures of Accuracy, Precision, Recall, and F-measures.

• International Journal of Computer Science & Network Security
• /
• 제24권5호
• /
• pp.111-118
• /
• 2024

In general network-based intrusion detection system is designed to detect malicious behavior directed at a network or its resources. The key goal of this paper is to look at network data and identify whether it is normal traffic data or anomaly traffic data specifically for accounting information systems. In today's world, there are a variety of principles for detecting various forms of network-based intrusion. In this paper, we are using supervised machine learning techniques. Classification models are used to train and validate data. Using these algorithms we are training the system using a training dataset then we use this trained system to detect intrusion from the testing dataset. In our proposed method, we will detect whether the network data is normal or an anomaly. Using this method we can avoid unauthorized activity on the network and systems under that network. The Decision Tree and K-Nearest Neighbor are applied to the proposed model to classify abnormal to normal behaviors of network traffic data. In addition to that, Logistic Regression Classifier and Support Vector Classification algorithms are used in our model to support proposed concepts. Furthermore, a feature selection method is used to collect valuable information from the dataset to enhance the efficiency of the proposed approach. Random Forest machine learning algorithm is used, which assists the system to identify crucial aspects and focus on them rather than all the features them. The experimental findings revealed that the suggested method for network intrusion detection has a neglected false alarm rate, with the accuracy of the result expected to be between 95% and 100%. As a result of the high precision rate, this concept can be used to detect network data intrusion and prevent vulnerabilities on the network.