• 한국정보통신학회논문지
• /
• 제25권12호
• /
• pp.1809-1816
• /
• 2021

국가연구개발사업 관련 데이터를 이용한 최신 연구동향 파악, 의미 있는 정보의 생산과 활용을 위해 국가R&D 정보 서비스에도 자동 분류 기술 적용이 요구되어 R&D과제의 연구분야를 자동 분류하고 추천하기 위한 연구를 진행했다. 2013~2020년 국가R&D 과제 데이터 약 45만 건을 수집하여 학습과 평가에 사용했다. 수집 데이터 중 유효한 데이터를 대상으로 데이터 전처리 및 분석, 실험을 통한 성능 분석 후 모델을 선정했다. 최적의 모델 조합 도출을 목적으로 Word2vec, GloVe, fastText 성능을 비교했다. 실험 결과, 과제정보의 필수 항목으로 사용되는 소분류만의 정확도는 90.11%이다. 이 모델은 국가과학기술표준분류 연구분야와 유사한 계층 구조를 가진 다른 분류체계의 자동 분류 연구에 활용 가능할 것으로 기대한다.

• ETRI Journal
• /
• 제39권1호
• /
• pp.87-96
• /
• 2017

Ultrasonic surgical devices are routinely used for surgical procedures. The incision and coagulation of tissue generate a temperature of $40^{\circ}C-150^{\circ}C$ and depend on the controllable output power level of the surgical device. Recently, research on the classification of grasped tissues to automatically control the power level was published. However, this research did not consider the specific characteristics of the surgical device, tissue denaturalization, and so on. Therefore, this research proposes a robust algorithm that simulates noise to resemble real situations and classifies tissue using conventional classifier algorithms. In this research, the bioimpedance spectrum for six tissues (liver, large intestine, kidney, lung, muscle, and fat) is measured, and five classifier algorithms are used. A signal-to-noise ratio of additive white Gaussian noise diversifies the testing sets, and as a result, each classifier's performance exhibits a difference. The k-nearest neighbors algorithm shows the highest classification rate of 92.09% (p < 0.01) and a standard deviation of 1.92%, which confirms high reproducibility.

• ETRI Journal
• /
• 제37권3호
• /
• pp.541-550
• /
• 2015

In this paper, we propose a classification-based approach for hybridizing statistical machine translation and rulebased machine translation. Both the training dataset used in the learning of our proposed classifier and our feature extraction method affect the hybridization quality. To create one such training dataset, a previous approach used auto-evaluation metrics to determine from a set of component machine translation (MT) systems which gave the more accurate translation (by a comparative method). Once this had been determined, the most accurate translation was then labelled in such a way so as to indicate the MT system from which it came. In this previous approach, when the metric evaluation scores were low, there existed a high level of uncertainty as to which of the component MT systems was actually producing the better translation. To relax such uncertainty or error in classification, we propose an alternative approach to such labeling; that is, a cut-off method. In our experiments, using the aforementioned cut-off method in our proposed classifier, we managed to achieve a translation accuracy of 81.5% - a 5.0% improvement over existing methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권9호
• /
• pp.2927-2941
• /
• 2022

The primary task of machine fault diagnosis is to judge whether the current state is normal or damaged, so it is a typical binary classification problem with mutual exclusion. Mutually exclusive events and out-of-domain detection have one thing in common: there are two types of data and no intersection. We proposed a fusion model method to improve the accuracy of machine fault diagnosis, which is based on the mutual exclusivity of events and the commonality of out-of-distribution detection, and finally generalized to all binary classification problems. It is reported that the performance of a convolutional neural network (CNN) will decrease as the recognition type increases, so the variational auto-encoder (VAE) is used as the primary model. Two VAE models are used to train the machine's normal and fault sound data. Two reconstruction probabilities will be obtained during the test. The smaller value is transformed into a correction value of another value according to the mutually exclusive characteristics. Finally, the classification result is obtained according to the fusion algorithm. Filtering normal data features from fault data features is proposed, which shields the interference and makes the fault features more prominent. We confirm that good performance improvements have been achieved in the machine fault detection data set, and the results are better than most mainstream models.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제12권4호
• /
• pp.219-228
• /
• 2006

최근 여러 생물학 연구 집단들은 연구의 효율 향상을 위해 그들의 연구 결과를 서로 공유하기 위한 노력을 하고 있다. 뿐만 아니라, 공통의 어휘를 이용하여 유전자의 기능을 기술하기 위해 통제된 어휘들로 이루어진 Gene Ontology(GO) 라는 온툴로지를 구축하였다. 하지만 현재까지도 각 연구 집단들의 데이타는 분산되어 있고, 기존의 시스템들은 이처럼 분산된 데이타들에 대한 통합 질의를 지원하지 않고 있을 뿐 아니라, 각 연구 집단의 독자적인 어휘들과 GO 와의 대응 관계에 대한 의미가 명확하게 기술되어 있지 않아 통합 시맨틱 질의가 근본적으로 불가능한 상태이다. 본 논문에서는 대응 관계의 의미를 결정하는 기법과, 통합 시맨틱 질의를 지원하는 인터페이스를 제안하였다. 먼저, 문자열 규칙과 다중도 분석 등을 통해 이러한 대응 관계의 의미를 반자동으로 결정해 주고 이렇게 결정된 대응 관계의 의미를 GO 와 통합하여 통합 온톨로지를 생성해 주는 AutoGOA 시스템을 제안하였다. 또한, 대표적인 메타데이타 기술 모델인 RDF 모델을 이용하여 여러 데이타들을 통합하고 이렇게 생성된 통합 온툴로지를 이용하여 통합 시맨틱 질의를 지원하는 인터페이스인 GOGuide II 를 제안하였다.

• 한국컴퓨터정보학회논문지
• /
• 제28권3호
• /
• pp.35-43
• /
• 2023

운전 중 감정 인식은 사고를 예방하기 위해 꼭 필요한 과제이다. 더 나아가 자율 주행 시대에서 자동차는 모빌리티의 주체로 운전자와의 감정적인 소통이 더욱 요구되고 있으며 감정 인식 시장은 점점 확산되고 있다. 이에 따라 본 연구 방안에서는 수집하기 비교적 용이한 데이터인 심리데이터와 행동 데이터를 이용해 운전자의 감정을 분류하는 인공지능 모델을 개발하고자 한다. 오토인코더 모델을 통해 잠재 변수를 추출하고, 이를 본 분류 모델의 변수로 사용하였으며, 이는 성능 향상에 영향을 미침을 확인하였다. 또한 기존 뇌파 데이터를 포함했을 때 보다 본 논문이 제시하는 프레임워크를 사용하였을 때 성능이 향상됨도 확인하였다. 최종적으로 심리 및 개인정보데이터, 행동 데이터만을 통해 운전자의 감정 분류 정확도 81%와 F1-Score 80%를 달성하였다.

• 대한의용생체공학회:의공학회지
• /
• 제41권2호
• /
• pp.101-106
• /
• 2020

Although benign gastric ulcers do not develop into gastric cancer, they are similar to early gastric cancer and difficult to distinguish. This may lead to misconsider early gastric cancer as gastric ulcer while diagnosing. Since gastric cancer does not have any special symptoms until discovered, it is important to detect gastric ulcers by early gastroscopy to prevent the gastric cancer. Therefore, we developed a Convolution Neural Network (CNN) model that can be helpful for endoscopy. 3,015 images of gastroscopy of patients undergoing endoscopy at Gachon University Gil Hospital were used in this study. Using ResNet-50, three models were developed to classify normal and gastric ulcers, normal and gastric cancer, and gastric ulcer and gastric cancer. We applied the data augmentation technique to increase the number of training data and examined the effect on accuracy by varying the multiples. The accuracy of each model with the highest performance are as follows. The accuracy of normal and gastric ulcer classification model was 95.11% when the data were increased 15 times, the accuracy of normal and gastric cancer classification model was 98.28% when 15 times increased likewise, and 5 times increased data in gastric ulcer and gastric cancer classification model yielded 87.89%. We will collect additional specific shape of gastric ulcer and cancer data and will apply various image processing techniques for visual enhancement. Models that classify normal and lesion, which showed relatively high accuracy, will be re-learned through optimal parameter search.

• 감성과학
• /
• 제14권2호
• /
• pp.293-300
• /
• 2011

CGI(Computer-generated Image) 기술은 이미지 생성을 자동화한다는 측면에서 디자이너에게 도움을 줄 수 있다. CGI를 활용하는 과정에서 두 가지 중요한 활동은 이미지의 제작과 관리이다. 다양성을 추구하는 디자이너에게 추상이미지의 자동 생성과 같은 기법은 자유로운 형태의 이미지 획득에 많은 도움을 줄 수 있다. 이와 더불어 중요한 이슈는 방대한 분량의 이미지를 적절한 메커니즘을 통해 관리하는 것이다. 추상이미지는 특성상 이미지에 대응하는 검색어 설정이 어려우며, 분류 역시 까다로운 문제로 남아있다. 본 논문은 디자이너에게 친숙한 조형 요소와 감성 요소를 분류와 표현의 정보로 이용함으로써 자동 생산된 추상 이미지를 효과적으로 분류하고 표현하는 방법을 제안한다. 추상이미지에 대한 적절한 분류와 표현은 이미지 데이타베이스 구축 및 검색에 있어 간결하고 효과적인 기술로 활용할 수 있으며 대규모 컨텐츠 관리에 도움을 줄 수 있다.

• 자동차안전학회지
• /
• 제10권2호
• /
• pp.13-19
• /
• 2018

The purpose of this study was to evaluate the effect of air bag deployment in passenger car head-on collisions on injuries to the driver. The drivers in head-on collisions who were brought to the emergency rooms of two hospitals from January 2011 and October 2014 were evaluated, as were the vehicles involved. The driver injury level were assessed by utilizing Collision Deformation Classification (CDC) codes, and the Abbreviated Injury Scale (AIS) and Injury Severity Score (ISS), respectively. In this study, it was shown that the chest ISS and AIS were significantly high when an air bag only is deployed. A statistically significant difference was found in the crush extent when the driver who fastened the seatbelt was found to be affected more than the ISS 9. Even when an air bag is deployed in a head-on car collision, injury severity can vary according to accident circumstances and crash severity. Accordingly, first aid can be rapidly given, and the injured person can be quickly referred to a hospital, only if the assessment of persons involved in a vehicle accident is accurately carried out.

• 한국정밀공학회지
• /
• 제16권2호통권95호
• /
• pp.87-94
• /
• 1999

The automated die design program is developed for cold forging of axisymmetric parts which are mainly produced by forward extrusion, backward extrusion, composite extrusion and upsetting. For this study, firstly classification of forged parts and investigation of die construction type usually employed in forging industry are carried out and the most proper type from several kinds of die construction is proposed as a standardized one. Based on the die design rules summarized in the references such as handbooks, technical papers, monthly journals, etc. the automated die design program was made using AutoLISP language available in AutoCAD software of personal computer. This program interactively runs for only input data, for example, forging process, shape of forged parts, type of punch, split of die insert and design of shrinkage rings and then displays details of drawings necessary to make a forging die. When a variety of forging processes and forged parts are tested to examine the validity of this program, it was confirmed to give good results applicable to the forging die design in press shop.