• 한국기계가공학회지
• /
• 제19권7호
• /
• pp.7-15
• /
• 2020

Recently, the fourth industrial revolution has been presented as a new paradigm and additive manufacturing (AM) has become one of the most important topics. For this reason, process monitoring for each cross-sectional layer of additive metal manufacturing is important. Particularly, deep learning can train a machine to analyze, optimize, and repair defects. In this paper, image classification is proposed by learning images of defects in the metal cross sections using the convolution neural network (CNN) image labeling algorithm. Defects were classified into three categories: crack, porosity, and hole. To overcome a lack-of-data problem, the amount of learning data was augmented using a data augmentation algorithm. This augmentation algorithm can transform an image to 180 images, increasing the learning accuracy. The number of training and validation images was 25,920 (80 %) and 6,480 (20 %), respectively. An optimized case with a combination of fully connected layers, an optimizer, and a loss function, showed that the model accuracy was 99.7 % and had a success rate of 97.8 % for 180 test images. In conclusion, image labeling was successfully performed and it is expected to be applied to automated AM process inspection and repair systems in the future.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.35.3-35
• /
• 2002

Reinforcement Learning (RL) is one of machine learning methods and an RL agent autonomously learns the action selection policy by interactions with its environment. At the beginning of RL research, it was limited to problems in environments assumed to be Markovian Decision Process (MDP). However in practical problems, the agent suffers from the incomplete perception, i.e., the agent observes the state of the environments, but these observations include incomplete information of the state. This problem is formally modeled by Partially Observable MDP (POMDP). One of the possible approaches to POMDPS is to use historical nformation to estimate states. The problem of these approaches is how t..

• 디지털산업정보학회논문지
• /
• 제14권4호
• /
• pp.69-77
• /
• 2018

The malignant melanoma accounts for about 1 to 3% of the total malignant tumor in the West, especially in the US, it is a disease that causes more than 9,000 deaths each year. Generally, skin lesions are difficult to detect the features through photography. In this paper, we propose a computer-aided diagnosis algorithm based on deep learning for classification of malignant melanoma and benign skin tumor in RGB channel skin images. The proposed deep learning model configures the tumor lesion segmentation model and a classification model of malignant melanoma. First, U-Net was used to segment a skin lesion area in the dermoscopic image. We could implement algorithms to classify malignant melanoma and benign tumor using skin lesion image and results of expert's labeling in ResNet. The U-Net model obtained a dice similarity coefficient of 83.45% compared with results of expert's labeling. The classification accuracy of malignant melanoma obtained the 83.06%. As the result, it is expected that the proposed artificial intelligence algorithm will utilize as a computer-aided diagnosis algorithm and help to detect malignant melanoma at an early stage.

• Nuclear Engineering and Technology
• /
• 제55권11호
• /
• pp.4057-4065
• /
• 2023

Organic scintillators are widely used for neutron/gamma detection. Pulse shape discrimination algorithms have been commonly used to discriminate the detected radiations. These algorithms have several limits, in particular with plastic scintillator which has lower discrimination ability, compared to liquid scintillator. Recently, machine learning (ML) models have been explored to enhance discrimination performance. Nevertheless, obtaining an accurate ML model or evaluating any discrimination approach requires a reference neutron dataset. The preparation of this is challenging because neutron sources are also gamma-ray emitters. Therefore, this paper proposes a pipeline to prepare clean labeled neutron/gamma datasets acquired by an organic scintillator. The method is mainly based on a Time of Flight setup and Tail-to-Total integral ratio (TTT_ratio) discrimination algorithm. In the presented case, EJ276 plastic scintillator and ²⁵²Cf source were used to implement the acquisition chain. The results showed that this process can identify and remove mislabeled samples in the entire ToF spectrum, including those that contribute to peak values. Furthermore, the process cleans ToF dataset from pile-up events, which can significantly impact experimental results and the conclusions extracted from them.

• 한국차세대컴퓨팅학회논문지
• /
• 제15권3호
• /
• pp.16-24
• /
• 2019

최근 뇌파를 기반으로 한 인간의 감정을 인식하는 연구가 인간-로봇 상호작용 분야에서 활발히 진행되고 있다. 본 논문에서는 MAHNOB-HCI에서 사용된 자기평가와 주석 레이블링 방법과는 다른, 이미지 기반의 뇌파 Topography를 이용한 레이블링을 통해 감정을 평가하는 방법을 제안한다. 제안한 방법은 뇌파 신호를 Topography의 이미지로 변환하여 기계학습 모델을 학습하고 이를 기반으로 Valence 기반의 감정을 평가한다. 제안한 방법은 레이블링 과정을 자동화하여 지연 시간을 없애고 객관적인 레이블링을 제공할 수 있다. MAHNOB-HCI 데이터베이스를 적용한 실험에서 SVM, kNN의 기계학습 모델을 학습하여 주석 레이블링과 성능 비교를 하였으며, 제안 방법의 감정인식 정확도를 SVM에서 54.2%, kNN에서 57.7%로 확인하였다.

• 정보과학회 논문지
• /
• 제44권7호
• /
• pp.686-691
• /
• 2017

텔레마케팅은 지식정보화 사회가 되면서 기업 마케팅 활동의 중심축으로 발전하였다. 최근 금융 데이터에 기계학습을 적용하는 연구가 활발하게 진행되고 있으며 좋은 성과를 내고 있다. 하지만 지도학습법이 대부분이어서 많은 양의 클래스가 있는 데이터가 필요하다. 본 논문에서는 텔레마케팅의 목표 고객을 선정하는데 클래스가 없는 금융 데이터에 자동으로 클래스를 부여하는 방법을 제안한다. 준지도 학습법 중 레이블 전파와 의사결정나무 기반의 협동 학습으로 클래스가 없는 데이터를 레이블링한다. 신뢰도가 낮은 데이터를 제거한 후 두 방법이 같은 클래스로 예측한 데이터만 추출한다. 이를 학습 데이터에 추가한 후 의사결정나무를 학습하여 테스트 데이터로 평가한다. 제안하는 방법의 유용성을 입증하기 위해 실제 포르투갈 은행의 텔레마케팅 데이터를 이용하여 실험을 수행하였다. 비교 실험 결과, 정확도가 83.39%로 1.82% 향상되고, 정밀도가 19.37%로 2.67% 향상되었으며, t-검증을 통해 유의미한 성능 향상이 있음을 입증하였다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권3호
• /
• pp.209-215
• /
• 2014

Sequence tagging is the task of predicting frame-wise labels for a given input sequence and has important applications to diverse domains. Conventional methods such as maximum likelihood (ML) learning matches global features in empirical and model distributions, rather than local features, which directly translates into frame-wise prediction errors. Recent probabilistic sequence models such as conditional random fields (CRFs) have achieved great success in a variety of situations. In this paper, we introduce a novel discriminative CRF learning algorithm to minimize local feature mismatches. Unlike overall data fitting originating from global feature matching in ML learning, our approach reduces the total error over all frames in a sequence. We also provide an efficient gradient-based learning method via gradient forward-backward recursion, which requires the same computational complexity as ML learning. For several real-world sequence tagging problems, we empirically demonstrate that the proposed learning algorithm achieves significantly more accurate prediction performance than standard estimators.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.338-340
• /
• 2019

최근 네트워크의 발달로 인해 데이터가 축적되는 속도와 크기가 증가되고 있다. 이 데이터들을 분류하는데 많은 어려움이 있는데 그 어려움 중에 하나가 라벨링의 어려움이다. 라벨링은 보통 사람이 진행하게 되는데 모든 사람이 같은 방식으로 데이터를 이해를 하는데 무리가 있어 동일한 기준으로 라벨링하는 것은 매우 어렵다는 문제가 있다. 이를 해결하기 위해 본 논문에서는 GAN을 이용하여 입력 이미지를 기반으로 새로운 이미지를 생성하고 이를 학습을 하는 데 사용을 하여 입력 데이터를 간접적으로 학습할 수 있게 구현하였다. 이를 통해 학습 데이터의 개수를 늘려 분류의 정확도를 높일 수 있을 것으로 사료된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권9호
• /
• pp.4814-4832
• /
• 2019

Plagiarism detection is increasingly exploiting text alignment. Text alignment involves extracting the plagiarism passages in a pair of the suspicious document and its source document. The heuristics have achieved excellent performance in text alignment. However, the further improvements of the heuristic methods mainly depends more on the experiences of experts, which makes the heuristics lack of the abilities for continuous improvements. To address this problem, machine learning maybe a proper way. Considering the position relations and the context of text segments pairs, we formalize the text alignment task as a problem of sequence labeling, improving the current methods at the model level. Especially, this paper proposes to use the probabilistic graphical model to tag the observed sequence of pairs of text segments. Hence we present the sequence labeling approach for text alignment in plagiarism detection based on Conditional Random Fields. The proposed approach is evaluated on the PAN@CLEF 2012 artificial high obfuscation plagiarism corpus and the simulated paraphrase plagiarism corpus, and compared with the methods achieved the best performance in PAN@CLEF 2012, 2013 and 2014. Experimental results demonstrate that the proposed approach significantly outperforms the state of the art methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2012-2030
• /
• 2021

Named Entity Recognition (NER) is used to identify entity nouns in the corpus such as Location, Person and Organization, etc. NER is also an important basic of research in various natural language fields. The processing of Chinese NER has some unique difficulties, for example, there is no obvious segmentation boundary between each Chinese character in a Chinese sentence. The Chinese NER task is often combined with Chinese word segmentation, and so on. In response to these problems, we summarize the recognition methods of Chinese NER. In this review, we first introduce the sequence labeling system and evaluation metrics of NER. Then, we divide Chinese NER methods into rule-based methods, statistics-based machine learning methods and deep learning-based methods. Subsequently, we analyze in detail the model framework based on deep learning and the typical Chinese NER methods. Finally, we put forward the current challenges and future research directions of Chinese NER technology.